Water Chemistry Effects On Dissolution Rates of Lead Corrosion Products

Water Chemistry Effects
on Dissolution Rates of
Lead Corrosion Products
Subject Area: Infrastructure

©2010 Water Research Foundation. ALL RIGHTS RESERVED

About the Water Research Foundation
The Water Research Foundation (formerly Awwa Research Foundation or AwwaRF) is a member-supported,
international, 501(c)3 nonprofit organization that sponsors research to enable water utilities, public health
agencies, and other professionals to provide safe and affordable drinking water to consumers.
The Foundation’s mission is to advance the science of water to improve the quality of life. To achieve this
mission, the Foundation sponsors studies on all aspects of drinking water, including resources, treatment,
distribution, and health effects. Funding for research is provided primarily by subscription payments from
close to 1,000 water utilities, consulting firms, and manufacturers in North America and abroad. Additional
funding comes from collaborative partnerships with other national and international organizations and the
U.S. federal government, allowing for resources to be leveraged, expertise to be shared, and broad-based
knowledge to be developed and disseminated.
From its headquarters in Denver, Colorado, the Foundation’s staff directs and supports the efforts of
more than 800 volunteers who serve on the board of trustees and various committees. These volunteers
represent many facets of the water industry, and contribute their expertise to select and monitor research
studies that benefit the entire drinking water community.
The results of research are disseminated through a number of channels, including reports, the Web site,
Webcasts, conferences, and periodicals.
For its subscribers, the Foundation serves as a cooperative program in which water suppliers unite to pool
their resources. By applying Foundation research findings, these water suppliers can save substantial costs
and stay on the leading edge of drinking water science and technology. Since its inception, the Foundation
has supplied the water community with more than $460 million in applied research value.
More information about the Foundation and how to become a subscriber is available on the Web at
www.WaterResearchFoundation.org.

Prepared by:
Daniel E. Giammar, Katherine S. Nelson, James D. Noel, and Yanjiao Xie
Department of Energy, Environmental and Chemical Engineering,
Washington University, 1 Brookings Drive, St. Louis, MO 63130
Sponsored by:
Water Research Foundation
6666 West Quincy Avenue, Denver, CO 80235-3098
Published by:

DISCLAIMER
This study was funded by the Water Research Foundation (Foundation). The Foundation
assumes no responsibility for the content of the research study reported in this publication
or for the opinions or statements of fact expressed in the report. The mention of trade names
for commercial products does not represent or imply the approval or endorsement of the
Foundation. This report is presented solely for informational purposes.
Copyright © 2010
by Water Research Foundation
ALL RIGHTS RESERVED.

No part of this publication may be copied, reproduced
or otherwise utilized without permission.
ISBN 978-1-60573-087-5
Printed in the U.S.A.

CONTENTS
LIST OF TABLES�� vii
LIST OF FIGURES�� ix
FOREWORD�� xv
ACKNOWLEDGMENTS�� xvii
EXECUTIVE SUMMARY�� xix
CHAPTER 1: INTRODUCTION�� 1
Significance of Lead Corrosion to Drinking Water Quality ............................................... 1
Project Objectives and Overview of Research Approach.................................................... 2
Project Objectives�� 2
Overview of Research Approach and Structure of the Report�� 3
Background on Lead Corrosion and Corrosion Control...................................................... 3
Lead Corrosion�� 3
Corrosion Products�� 4
Corrosion Control�� 5
Lead Release Rates From Pipe�� 6
CHAPTER 2: REVIEW OF LEAD CARBONATE, PHOSPHATE, AND OXIDE

DISSOLUTION RATES AND SOLUBILITY�� 7
Approach to the Review....................................................................................................... 7
Summary of the Review....................................................................................................... 7
Lead Solubility..................................................................................................................... 7
Dissolution Rates of Precipitated Solids�� 10
CHAPTER 3: DISSOLUTION RATES OF LEAD CORROSION PRODUCTS�� 13

Overview of Research on Dissolution Rates..................................................................... 13
Materials and Methods....................................................................................................... 13
Preparation and Characterization of Lead Corrosion Products�� 13
Measurement of Dissolution Rates �� 18
Results and Discussion...................................................................................................... 22
Hydrocerussite Dissolution Rates�� 22
Plattnerite Dissolution Rates�� 34
Hydroxylpyromorphite Dissolution Rates�� 49
Summary of Rate Measurements and Limitations of Present Study�� 63

vi | Water Chemistry Effects on Dissolution Rates of Lead Corrosion Products
CHAPTER 4: RELEASE OF LEAD FROM PIPE SCALES�� 65

Overview of Research on Lead Release From Pipe Scales............................................... 65
Materials and Methods....................................................................................................... 65
Lead Pipes With Scales Containing Corrosion Products�� 65
Pipe Release Experiments�� 68
Results and Discussion...................................................................................................... 71
Evolution of Water Composition and Pipe Scales During Conditioning
of New Lead Pipe�� 71
Lead Release From Laboratory-Conditioned Pipes�� 76
Lead Release From Pipe Sections of Lead Service Lines�� 83
CHAPTER 5: SUMMARY AND CONCLUSIONS�� 93

Summary of Project........................................................................................................... 93
Dissolution Rates of Lead Corrosion Products.................................................................. 93
Dissolution Rate of Hydrocerussite�� 93
Dissolution Rate of Plattnerite�� 94
Dissolution Rate of Hydroxylpyromorphite�� 94
Lead Release From Scales on Lead Pipes.......................................................................... 95
Corrosion Products Present in Pipe Scales�� 95
Lead Release From Pipe Sections�� 95
CHAPTER 6: RECOMMENDATIONS TO UTILITIES�� 97

Importance of Collecting Data on Water and Solids.......................................................... 97
Potential Processes Impacting Corrosion Control............................................................. 97
APPENDIX A: REVIEW OF LEAD CARBONATE, PHOSPHATE, AND OXIDE

DISSOLUTION RATES�� 99
APPENDIX B: EQUILIBRIUM CONSTANTS AND REACTIONS�� 113
APPENDIX C: RELATIONSHIP AMONG ALKALINITY, pH, AND DISSOLVED

INORGANIC CARBON�� 117
APPENDIX D: TRACER STUDY OF CONTINUOUS-FLOW STIRRED REACTOR�� 119
APPENDIX E: CALCULATING EQUILIBRIUM LEAD CONCENTRATIONS FOR

HYDROCERUSSITE EXPERIMENTS�� 121
APPENDIX F: ORDER OF CONDUCTING EXPERIMENTS WITH LEAD PIPES �� 125
APPENDIX G: RESULTS OF EXPERIMENTS WITH LEAD SERVICE LINES�� 127
REFERENCES�� 131
ABBREVIATIONS�� 137

TABLES
3.1 Solid phases studied in dissolution rate experiments......................................................... 14
3.2 Experimental variables in systematic investigation of dissolution rates........................... 19
3.3 Conditions studied for hydrocerussite dissolution............................................................. 22
3.4 Conditions and results of hydrocerussite dissolution experiments.................................... 23
3.5 Conditions studied for plattnerite dissolution.................................................................... 35
3.6 Conditions and results of plattnerite dissolution experiments........................................... 36
3.7 Conditions studied for hydroxylpyromorphite dissolution................................................ 49
3.8 Conditions and results of hydroxylpyromorphite dissolution experiments....................... 50
4.1 List of lead pipe supplied by the Massachusetts Water Resources Authority
from one of its customer communities............................................................................... 66
4.2 Factors evaluated in experiments with pipe reactors......................................................... 70
4.3 Water compositions evaluated in lead release experiments with pipe sections................. 71
4.4 Dissolved lead divided by total lead concentration........................................................... 83
A.1 Equilibrium constants and reactions for dissolved lead species included
in the calculation of total dissolved lead concentrations................................................. 102
A.2 Summary of selected previous studies of carbonate dissolution rates............................. 109
A.3 Summary of selected previous studies of phosphate dissolution rates............................ 110
B.1 Equilibrium constants for aqueous species...................................................................... 114
B.2 Chemical potentials for various aqueous species............................................................. 115
B.3 Equilibrium constants for select reactions of aqueous species........................................ 115
B.4 Solubility product of select lead solids............................................................................ 116
C.1 Alkalinity for combinations of pH and DIC investigated................................................ 117
vii

viii | Water Chemistry Effects on Dissolution Rates of Lead Corrosion Products
E.1 Equilibrium constants for aqueous species...................................................................... 122
F.1 Schedule of experiments with laboratory-developed scales in new lead pipes............... 125
F.2 Order of experiments with pipe sections from an actual lead service line...................... 126
G.1 Conditions and results of lead release experiments......................................................... 127

FIGURES
ES.1 Formation and stability of lead-containing corrosion products in the scales

of lead pipes...................................................................................................................... xx
1.1 Formation and stability of lead-containing corrosion products in the scales

of lead pipes........................................................................................................................ 2
1.2 Overview of research approach.......................................................................................... 3
2.1 Lead solubility as a function of pH..................................................................................... 8
2.2 Predominance area diagrams showing the dominant lead solid phase
of dissolved species as a function pH and oxidation-reduction potential........................... 9
3.1 XRD pattern of hydrocerussite and reference pattern....................................................... 14
3.2 Synthesized hydrocerussite at two different levels of magnification................................ 15
3.3 Synthesized cerussite........................................................................................................ 15
3.4 X-ray diffraction patterns of as-synthesized and acid-washed hydroxylpyromorphite.... 16
3.5 Acid-washed hydroxylpyromorhophite............................................................................ 16
3.6 Scanning electron micrograph of plattnerite..................................................................... 17
3.7 X-ray diffraction patterns of plattnerite............................................................................ 17
3.8 Flow-through reactor for measuring dissolution rates...................................................... 19
3.9 Illustration of the dissolved lead effluent from a flow-through reactor............................ 20
3.10 The effect of DIC on effluent lead concentrations from flow-through reactors
loaded with hydrocerussite at pH = 7.5............................................................................ 26
3.11 Dissolution rates of hydrocerussite as a function of pH with influent

concentrations of dissolved inorganic carbon (0, 10, and 50 mg C/L)............................. 27
3.12 Predicted equilibrium dissolved lead concentrations (lines) of 1 g/L hydrocerussite

as a function of pH and dissolved inorganic carbon concentration.................................. 28
3.13 The reaction rate constant (mol/min·m2) plotted as a function of pH and influent
dissolved inorganic concentration..................................................................................... 29
ix

x | Water Chemistry Effects on Dissolution Rates of Lead Corrosion Products
3.14 The reaction rate constant (mol/min·m2) plotted as a function of carbonate (CO32–)
concentration and influent dissolved inorganic concentration.......................................... 29
3.15 Dissolution rates of hydrocerussite as a function of pH at a DIC = 0 mg/L with

and without 4 mg P/L orthophosphate.............................................................................. 30
3.16 Dissolution rates of hydrocerussite as a function of pH and dissolved inorganic

carbon with an influent orthophosphate concentration of 4 mg P/L................................. 30
3.17 Powder X-ray diffraction of hydrocerussite reacted with an influent concentration

of pH = 8.5, DIC = 0 mg/L, and a P = 4 mg/L.................................................................. 31
3.18 Scanning electron microscopy image of (A) hydrocerussite reacted at pH = 10

and DIC = 10 mg/L and (B) hydrocerussite reacted with a pH = 8.5 and P = 4 mg/L
after 10 minutes in a batch reactor.................................................................................... 31
3.19 Predicted dissolved lead concentrations (lines) in equilibrium with hydroxyl-

pyromorphite with an influent orthophosphate concentration of 4 mg P/L...................... 32
3.20 Dissolution rates of hydrocerussite as a function of pH with and without 2 mg/L

as Cl2 monochloramine..................................................................................................... 33
3.21 Dissolution rates of hydrocerussite as a function of pH with and without 2 mg/L

as Cl2 monochloramine in the presence of 4 mg P/L orthophosphate and
10 mg C/L DIC................................................................................................................. 34
3.22 Plattnerite dissolution rate as a function of pH and DIC.................................................. 39
3.23 Effect of DIC on plattnerite dissolution rates................................................................... 40
3.24 Steady-state and predicted equilibrium concentrations for plattnerite dissolution

as a function of pH and DIC............................................................................................. 41
3.25 Conceptual model of PbO2 dissolution showing three different possible

mechanisms and the potential roles of carbonate and orthophosphate............................. 42
3.26 Effect of orthophosphate on dissolution rates of plattnerite............................................. 43
3.27 X-ray diffraction patterns of solids remaining following flow-through dissolution

experiments of plattnerite under conditions of 1 mg P/L orthophosphate, pH 8.5,
and DIC of 0 or 50 mg C/L............................................................................................... 44
3.28 Steady-state and predicted equilibrium concentrations for plattnerite dissolution

in the presence of 1 mg P/L orthophosphate..................................................................... 45
3.29 Effect of chloramines on dissolution rates of plattnerite.................................................. 46

Figures | xi
3.30 Empirical linear model of plattnerite dissolution rates..................................................... 47
3.31 Empirical exponential model of plattnerite dissolution rate constants............................. 48
3.32 Dissolution rates of hydroxylpyromorphite as a function of pH and orthophosphate

concentration..................................................................................................................... 53
3.33 Steady-state effluent and predicted equilibrium concentrations for

hydroxylpyromorphite dissolution.................................................................................... 54
3.34 Effect of dissolved inorganic carbon on hydroxylpyromorphite dissolution rates........... 55
3.35 Effect of monochloramine on hydroxylpyromorphite dissolution rates........................... 56
3.36 Comparison of steady-state effluent (open squares) and predicted equilibrium lead
concentrations for dissolution of hydroxylpyromorphite in the presence
of 2 mg Cl2/L of chloramines and 1 mg P/L of orthophosphate....................................... 57
3.37 Effects of combinations of water chemistry parameters on hydroxylpyromorphite

dissolution rates................................................................................................................ 58
3.38 X-ray diffraction patterns of hydroxylpyromorphite before and after reaction

at different pH and orthophosphate concentrations.......................................................... 59
3.39 Hydroxylpyromorphite after reacting with 2 mg Cl2/L of chloramine at pH 8.5............. 60
3.40 Hydroxylpyromorphite after reacting with 1 mg P/L of orthophosphate and

2 mg Cl2/L of chloramine at pH 8.5.................................................................................. 60
3.41 Hydroxylpyromorphite after reacting with 1 mg P/L of orthophosphate, 50 mg C/L

of dissolved inorganic carbon, and 2 mg Cl2/L of chloramine at pH 8.5.......................... 60
3.42 Hydroxylpyromorphite after reacting with 50 mg C/L of DIC and 2 mg Cl2/L of

chloramine at pH 8.5. (B) Abnormal large flat plate crystal, and (C) abnormal large
hexagonal crystal.............................................................................................................. 61
3.43 Dissolution rate constant, k (mol·m–2·min–1), of hydroxylpyromorphite as

a function of pH................................................................................................................ 61
4.1 Experimental pipe reactor system for studying lead release rates from pipe scales
following various stagnation periods and flow velocities................................................. 66
4.2 Mineralogy of pipe corrosion products of lead pipes removed from the MWRA
distribution system............................................................................................................ 67

xii | Water Chemistry Effects on Dissolution Rates of Lead Corrosion Products
4.3 System for evaluating lead release from lead pipes with scales developed in the
laboratory.......................................................................................................................... 69
4.4 Pipe reactor for investigating the lead release rates from pipe scales............................... 70
4.5 Lead concentration (a), pH (b), and residual chlorine concentration (c) in set
of four lead pipes (pipes P1–4) conditioned for 8 months................................................ 72
4.6 Lead concentration (a), pH (b), and residual chlorine concentration

(c) in set of nine lead pipes (pipes P5–13) conditioned for 3 months............................... 73
4.7 nXRD patterns of pipes scales formed after conditioning................................................ 74
4.8 Particles scraped off from pipe scales of Pipe 2 (a) and Pipe 13 (b)................................ 75
4.9 Cross section of pipe P2 surface with development of corrosion products...................... 76
4.10 Dissolved and total lead concentrations following 2 hours of reaction with three
different solutions in laboratory-conditioned lead pipes.................................................. 77
4.11 Change in pH following reaction for 2 hours with and without recirculating flow.......... 78
4.12 Effect of recirculating flow on dissolved and total lead concentrations........................... 78
4.13 Effect of flow on orthophosphate concentration............................................................... 79
4.14 Dissolved lead concentration over stagnation times following a 2 hour recirculation
period................................................................................................................................ 80
4.15 Total lead concentration over stagnation times following a 2 hour recirculation
period................................................................................................................................ 82
4.16 Dissolved and total lead concentrations following 2 hours of recirculation at

0.05 m/s velocity through lead pipes from MWRA.......................................................... 84
4.17 Effect of recirculating flow on (a) dissolved and (b) total lead concentrations
in MWRA pipe sections. Results are shown for the pipe that was first studied
at the low flow velocity condition..................................................................................... 86
4.18 Lead concentrations as a function of time for the low flow (0.05 m/s)
MWRA pipe experiments with influent conditions of pH = 8.5 and
DIC = 50 mg C/L (Solution 1).......................................................................................... 87
4.19 Lead concentrations as a function of time for the low flow (0.05 m/s)
MWRA pipe experiments with influent conditions of pH = 10 and
DIC = 10 mg C/L (Solution 3).......................................................................................... 88

Figures | xiii
4.20 Lead and orthophosphate concentrations as a function of time for the low
flow (0.05 m/s) MWRA pipe experiments with influent conditions
of pH = 7.5, DIC = 10 mg C/L, and P = 1 mg P/L (Solution 2)....................................... 90
A.1 Distribution of dissolved lead species as a function of pH for a solution

containing 15 μg/L of lead and 50 mg/L dissolved inorganic carbon............................. 103
A.2 Dissolved lead concentration and solid phases calculated to be present

for a system initially containing 1 g/L hydrocerussite.................................................... 105
A.3 Dissolved lead concentration and solid phases calculated to be present for a
system initially containing 1 g/L hydroxylpyromorphite, 10 mg/L dissolved
inorganic carbon, and (a) 1 mg/L or (b) 4 mg/L of dissolved orthophosphate............... 106
A.4 Predominance areas of lead(0)/lead(II)/lead(IV) system for 15 μgL total lead,

30 mg/L dissolved inorganic carbon, and (a) no orthophosphate or
(b) 3 mg/L orthophosphate.............................................................................................. 106
A.5 Dissolved lead concentration in equilibrium with lead(IV) oxide at

sufficiently oxidizing conditions that all lead is present as lead(IV).............................. 107
D.1 Dissolved lead concentration in a well-mixed flow-through reactor with a

30 minute hydraulic residence time................................................................................ 119
E.1 The effect of influent dissolved inorganic carbon on the total dissolved
lead concentration (Pbdiss) in equilibrium with hydrocerussite....................................... 123

FOREWORD
The Water Research Foundation (Foundation) is a nonprofit corporation that is dedicated

to the implementation of a research effort to help utilities respond to regulatory requirements
and traditional high-priority concerns of the industry. The research agenda is developed through
a process of consultation with subscribers and drinking water professionals. Under the umbrella
of a Strategic Research Plan, the Research Advisory Council prioritizes the suggested projects
based upon current and future needs, applicability, and past work; the recommendations are for-
warded to the Board of Trustees for final selection. The Foundation also sponsors research projects
through the unsolicited proposal process; the Collaborative Research, Research Applications, and
Tailored Collaboration programs; and various joint research efforts with organizations such as the
U.S. Environmental Protection Agency, the U.S. Bureau of Reclamation, and the Association of
California Water Agencies.
This publication is a result of one of these sponsored studies, and it is hoped that its find-
ings will be applied in communities throughout the world. The following report serves not only as
a means of communicating the results of the water industry’s centralized research program but also
as a tool to enlist the further support of the nonmember utilities and individuals.
Projects are managed closely from their inception to the final report by the Foundation’s
staff and large cadre of volunteers who willingly contribute their time and expertise. The Foundation
serves a planning and management function and awards contracts to other institutions such as water
utilities, universities, and engineering firms. The funding for this research effort comes primarily
from the Subscription Program, through which water utilities subscribe to the research program
and make an annual payment proportionate to the volume of water they deliver and consultants and
manufacturers subscribe based on their annual billings. The program offers a cost-effective and
fair method for funding research in the public interest.
A broad spectrum of water supply issues is addressed by the Foundation’s research agenda:
resources, treatment and operations, distribution and storage, water quality and analysis, toxicol-
ogy, economics, and management. The ultimate purpose of the coordinated effort is to assist water
suppliers to provide the highest possible quality of water economically and reliably. The true ben-
efits are realized when the results are implemented at the utility level. The Foundation’s trustees
are pleased to offer this publication as a contribution toward that end.
David E. Rager Robert C. Renner, P.E.

Chair, Board of Trustees Executive Director
Water Research Foundation Water Research Foundation
xv

ACKNOWLEDGMENTS
The authors of this report are indebted to the following water utilities and individuals for
their cooperation and participation in this project:
Leland Harms, Raymore, Mo.

Massachusetts Water Resources Authority, Boston, Mass., Windsor Sung
U.S. Environmental Protection Agency National Risk Management Research Laboratory,
Cincinnati, Ohio, Michael Schock
The authors appreciate the advice and guidance of the Project Advisory Committee
(PAC)—Jeff Bickel, AquaAmerica Pennsylvania; David Hokanson, Trussell Technologies, Inc.,
Pasadena, Calif.; and Heath Lloyd, City of Savannah Water and Sewer Department, Savannah,
Ga.—and the Water Research Foundation Project Manager, Traci Case. In the preparation of this
report, both Windsor Sung and Stephen Estes-Smargiassi of the Massachusetts Water Resources
Authority provided valuable comments.
The authors acknowledge the assistance of Tyler Nading, Vidhi Singhal, and Yin Wang in
conducting selected experiments and associated analyses.
xvii

EXECUTIVE SUMMARY
BACKGROUND AND APPROACH
Significance of Dissolution Rates of Lead Corrosion Products
The internal corrosion of lead-containing pipe, fittings, and solder in premise plumbing is
the most significant source of lead to drinking water. Although new construction does not use lead
pipe, many older buildings retain original lead service lines and internal plumbing. Concern over
lead concentrations in drinking water motivated the passage of the Lead and Copper Rule (LCR)
in 1991. The LCR requires utilities to implement methods to control lead corrosion if the 90th
percentile of samples exceeds the action level of 0.015 mg/L. Lead concentrations in tap water
are strongly influenced by distribution system water chemistry. In response to changes in water
chemistry, high lead concentrations can also be observed in systems with no previous history of
a lead problem. The increase in lead concentrations in the Washington D.C. Water and Sewer
Authority service area that occurred following a switch to chloramination for disinfection is a dra-
matic example of lead release that can result from changes in water chemistry.
Lead concentrations in drinking water are affected by chemical reactions that occur within
lead service lines and premise plumbing. Lead may be released directly from pipe, lead-containing
corrosion products on the pipe surface, and from brass and solder that contain lead (Figure ES.1).
Lead corrosion products observed on lead pipes include lead(II) carbonates and phosphates and,
more recently, lead(IV) oxide (PbO2(s)). Solubility and dissolution rates of corrosion products are
affected by water chemistry parameters including pH, dissolved inorganic carbon, orthophosphate,
and the concentration and type of disinfectant residual.
Information on dissolution rates of lead corrosion products can be particularly valuable as
water suppliers consider process changes that affect water chemistry such as switching disinfectant
type or dose, adjusting pH, or adding a corrosion inhibitor. While the formation of lead corrosion
products can passivate the pipe surface and limit further lead release, changes in water chemistry
can cause dissolution of the corrosion products and lead release to solution. An improved under-
standing of the effects of water chemistry on lead release rates can be used to identify strategies
that inhibit dissolution and minimize lead release. Information on corrosion product dissolution
rates can also help quantify the influence of stagnation time on lead concentrations to which con-
sumers are exposed and upon which monitoring protocols are based.
Project Objectives
The primary objective of the project was to provide new information to the water supply
community that advances understanding of lead corrosion product dissolution and transformation
rates. Specific objectives were to:
1. Summarize the state-of-knowledge regarding dissolution rates of lead corrosion

products.
xix

xx | Water Chemistry Effects on Dissolution Rates of Lead Corrosion Products
Figure ES.1 Formation and stability of lead-containing corrosion products in the scales of
lead pipes
2. Determine dissolution and transformation rates of lead corrosion products as a func-

tion of pH and concentrations of orthophosphate, dissolved inorganic carbon, and
chloramines.
3. Evaluate lead release rates from corrosion products on lead pipes by applying funda-
mental dissolution rate information developed for Objective 2.
Research Approach
The three objectives were addressed in a research approach that was divided into three
corresponding research tasks. Task 1 was a literature review regarding the dissolution rates of
lead(II) carbonate, lead(II) phosphate, and lead(IV) oxide precipitates. Task 2 was a systematic
experimental investigation of the dissolution rates of three important lead corrosion products. Task
3 extended the study of dissolution rates of pure solids to release rates from pipe scales. Tasks 2
and 3 involved bench-scale laboratory experiments with integrated analysis of the aqueous solu-
tions and the solid phases.
The review prepared for Task 1 is intended to be a resource for water supply professionals
and for engineers and scientists conducting corrosion research. By expanding the literature review
beyond the field of water supply and treatment, valuable information was summarized in a form
for the water supply community.
Dissolution rates were determined in Task 2 for three important lead corrosion products:
the lead carbonate hydrocerussite (Pb3(CO3)2(OH)2(s)), the lead phosphate hydroxylpyromorphite
(Pb5(PO4)3OH(s)), and the lead(IV) oxide plattnerite (PbO2(s)). The dissolution rates of these solids
were determined as a function of pH (7.5, 8.5, and 10.0), orthophosphate concentration (0, 1, and
4 mg P/L), dissolved inorganic carbon concentration (0, 10, and 50 mg C/L), and the presence or
absence of chloramines (2 mg/L as Cl2). The dissolution rates of the lead corrosion products were
quantified using bench-scale (84 mL) completely-mixed continuous-flow reactors. In addition to
monitoring lead release to quantify dissolution rates, the solid phases were characterized to evalu-
ate possible transformation of the solid phases. Characterization of the pristine and reacted materi-
als primarily involved surface area measurement, X-ray diffraction (XRD), and scanning electron
microscopy (SEM). The solution compositions were evaluated in a reaction-based framework to
assess the relative controls of the concentrations by equilibrium versus kinetic processes and to
explore the development of a general dissolution rate model.
Task 3 extended the work on dissolution rates of pure solid phases to the investigation
of lead release rates from corrosion products developed as components of scales on lead pipe.

Executive Summary | xxi
Two pipe scales were investigated, one in lead service lines provided by the Massachusetts Water
Resources Authority (MWRA) after 80–100 years of use and another that was developed in the
laboratory on new lead pipe exposed to a solution with free chlorine concentrations that promote
the development of corrosion products. The pipe scales were characterized by SEM and XRD to
identify the corrosion products present and the thickness of scales. Lead release from the pipe sec-
tions was then measured as a function of water chemistry, water flow velocity, and stagnation time.
The pipes were exposed to water compositions other than those at which the pipe scales had accli-
mated to examine the release of lead from the scales that can occur in response to a change in water
chemistry. Three different water compositions were evaluated that represent possible corrosion
management strategies: (1) increasing dissolved inorganic carbon or alkalinity (50 mg C/L DIC at
pH 8.5), (2) adding orthophosphate (1 mg P/L at pH 7.5 and DIC of 10 mg C/L), and (3) increasing
pH (pH 10 with 10 mg C/L DIC). No disinfectant residual was included in these water composi-
tions. The impact of laminar versus turbulent flow was assessed by conducting experiments at
two different flow velocities. The goal of experiments with no flow and with two velocities was to
assess the impact of flow rate on release rates. The effect of flow was assessed in experiments with
recirculation of water through sections of lead pipe, which provides much longer contact times of
pipe scales with flowing water than will occur with the once-through flow of pipes in service. Lead
release was monitored for stagnation times of up to 48 hours.
RESULTS AND CONCLUSIONS
Task 1: Review of Lead Carbonate, Phosphate, and Oxide Dissolution Rates
The review introduces the impact of the dissolution of lead corrosion products on drink-
ing water quality. It then presents an overview of the equilibrium solubility of lead(II) carbonates
and phosphates and lead(IV) oxides. The dissolution rates of these solids are then summarized
and compared with the rates of other divalent metal carbonate and phosphate precipitates. The
presentation of rates describes an approach that accounts for the effects of pH and the distance of
the solution from equilibrium solubility. The dissolution rates of both carbonate and phosphate
solids increase with decreasing pH; however, the majority of these studies observed such effects at
pH values below 7, and measurements over a neutral pH range are less pH-dependent. Dissolution
rates of lead carbonates and phosphates are comparable to those of the much more frequently stud-
ied calcium carbonates and phosphates. Dissolution rates are lower at lower temperatures.
Task 2: Quantification of Dissolution and Transformation Rates of Lead Corrosion

Products
Combinations of pH and concentrations of DIC, orthophosphate, and chloramines were

evaluated for all three of the lead corrosion products. In total, 23 conditions were evaluated for
hydrocerussite, 21 for hydroxylpyromorphite, and 21 for plattnerite. For the simplest aqueous
compositions (i.e., no DIC, orthophosphate, or chloramines), hydrocerussite had dissolution rates
that were about two orders of magnitude greater than those of plattnerite and hydroxylpyromor-
phite. The effects of specific water chemistry parameters and combinations of water chemistry
parameters were examined for each solid.

xxii | Water Chemistry Effects on Dissolution Rates of Lead Corrosion Products
Hydrocerussite Dissolution Rates
Hydrocerussite dissolution rates were affected by pH and the concentrations of DIC and
orthophosphate.
• Dissolution rates generally decreased with increasing pH.

• The addition of DIC acted in two different ways. DIC addition decreased the dis-
solution rate of hydrocerussite by lowering the equilibrium lead concentration but
increased the dissolution rate through the formation of soluble lead(II)-carbonate
complexes. The effect of lowering the solubility of hydrocerussite is more significant.
• Orthophosphate addition lowered the net release rate of lead by more than two orders
of magnitude. The decrease was caused by the formation of the lead phosphate solid
hydroxylpyromorphite, a phase that was directly observed by SEM and XRD.
• Steady-state effluent lead concentrations for hydrocerussite were very close to pre-
dicted equilibrium values. This suggests that dissolved lead concentrations were con-
trolled by the equilibrium and not the kinetics of hydrocerussite dissolution.
Plattnerite Dissolution Rates
Dissolution rates of plattnerite, which is only stable in the presence of a free chlorine
residual, provide information about the rates of lead release when the water contacting pipe scales
no longer contains free chlorine.
• The dissolution rate of plattnerite increased with decreasing pH and with increas-
ing DIC. The effect of DIC is consistent with a dissolution mechanism explained by
the formation of soluble lead(II)-carbonate complexes that accelerate the release of
lead(II) formed by reduction of lead(IV) on the plattnerite surface. While most steady-
state effluent concentrations for a 30 minute residence time were quite low, concentra-
tions would increase for longer contact times of plattnerite with the water.
• Addition of orthophosphate significantly decreased the net rate of lead release. The
decrease is likely caused by a combination of mechanisms of precipitation of the
lead(II) phosphate solid hydroxylpyromorphite and by adsorption of phosphate to
the plattnerite surface to block sites of dissolution.
• The presence of chloramines may also inhibit plattnerite dissolution. The magnitude
of this effect was lower than that of orthophosphate addition.
• Dissolved lead concentrations were controlled by the dissolution rate of plattnerite
and not by its equilibrium solubility in the absence of orthophosphate. When ortho-
phosphate but not DIC was present, then dissolved concentrations were consistent
with hydroxylpyromorphite solubility.
Hydroxylpyromorphite Dissolution Rates
The dissolution rates of hydroxylpyromorphite were affected by all four water chemistry
parameters.

Executive Summary | xxiii
• Increasing pH from 7.5 to 8.5 decreased the dissolution rate when no orthophosphate,
DIC, or chloramines were added.
• Dissolution rates were lowered by the addition of orthophosphate. Adding 4 mg P/L
provided little or no advantage over the addition of only 1 mg P/L.
• DIC can increase the rate of hydroxylpyromorphite dissolution. This effect is caused
by the complexation of lead(II) by the carbonate ion and is most significant at high
pH. Consequently, high carbonate alkalinity may actually decrease the overall stabil-
ity of hydroxylpyromorphite in pipe scales.
• Dissolution rates were lower in the presence of chloramines. However, the cause may
be the chloride ions that were present in the ammonium chloride solutions used to cre-
ate chloramines. Chloride can promote the formation of chloropyromorphite, a phase
that is less soluble than hydroxylpyromorphite when sufficient chloride is present.
• The steady-state concentrations were close to predicted equilibrium concentrations
for most of the experiments conducted when orthophosphate was present and DIC
was absent. When DIC concentrations are significant or contact times are short, then
knowledge of dissolution rates may be important in estimating dissolved lead concen-
trations in pipe sections that contain hydroxylpyromorphite.
Task 3: Evaluation of Lead Release From Pipe Scales
Characterization of Corrosion Products in Pipe Scales
The lead pipe scales used for release experiments were characterized before their use in
release rate experiments.
• The pipe sections provided by the MWRA from one of its customer communities con-
tained primarily hydrocerussite and some cerussite. The formation of these phases is
consistent with the water chemistry in the distribution system.
• For the new lead pipes conditioned in the laboratory, the dominant corrosion products
were hydrocerussite and litharge (PbO). Pipes conditioned for 8 months had minor
amounts of the lead(IV) oxides scrutinyite and plattnerite, but pipes conditioned for
3 months did not have any of these phases. The thickness of the corrosion product
layer was about 30 μm for the pipes conditioned for 8 months and about 15 μm for
3 months of conditioning. During pipe conditioning the dissolved lead concentrations
stabilized around 50 µg/L within 50 days, but the concentrations were much higher
in the initial phases of conditioning. This relatively high free chlorine concentration
of 3.5 mg/L as Cl2 was chosen to promote the development of lead(IV) oxides in pipe
scales. Even after lead concentrations stabilized, oxidation of the pipe was still signifi-
cant as indicated by the continued consumption of free chlorine.
Lead Release From Laboratory-Conditioned Pipes
Lead release from the laboratory-conditioned pipes was affected by the composition of the
solution in the pipes. Releases were measured at conditions that represent a change in water chem-
istry from that with which the pipe scales had acclimated.

xxiv | Water Chemistry Effects on Dissolution Rates of Lead Corrosion Products
• The solution with 1 mg P/L orthophosphate resulted in the lowest releases of lead to
solution. The next most effective solution was the solution with elevated pH (pH 10),
and the highest concentrations were observed for the solution with high DIC (50 mg
C/L) and moderate pH (pH 8.5). These trends are consistent with predicted equilib-
rium concentrations.
• Dissolved and total lead concentrations increased significantly when two of the three
solutions were recirculated through the pipes. The total lead increased with flow more
than the dissolved lead did, which indicates that the recirculating flow mobilized
some particulate lead. However, the mobilization of particles is not the only cause of
the higher lead concentrations. The flowing water may provide greater mixing of the
water in the pipes to overcome limitations of lead release caused by transport of dis-
solved lead away from the corrosion products on the pipe surface. Recirculation pro-
vides much longer contact times of lead corrosion products with flowing water than
will occur with single pass flow through lead service lines in use, and consequently
the lead concentrations can also be much higher.
Lead Release From Sections of Actual Lead Service Lines
The influence of water chemistry on lead release from the actual lead service line pipe
sections was similar to its influence on the release from the laboratory-conditioned pipes. The
pipes were reacted with solutions of different compositions than experienced when they were most
recently in use, and the pipe sections were used directly without any reacclimation period.
• Orthophosphate addition was the most effective of the three approaches for mitigat-
ing lead release. For the solutions without orthophosphate, the dissolved lead con-
centrations were consistent with measured concentrations from experiments with
pure hydrocerussite in Task 2. With continuous addition of orthophosphate, the solu-
tion with orthophosphate would likely be able to maintain the lowest dissolved lead
concentrations.
• Recirculating flow increased both dissolved and total lead concentrations. For one of
the solutions, the increase in flow rate also mobilized significant amounts of particu-
late lead. Contact times with flowing water were much greater than the pipe sections
would have experienced when in service, and consequently measured lead concentra-
tions are also much higher.
• Stagnation time had little effect on the concentrations of dissolved lead in the pipe
sections. For solutions without orthophosphate, dissolved lead concentrations reached
stable values within 1 hour. Consequently, many estimates of dissolved lead concen-
trations in hydrocerussite-coated pipe sections could be made based only on equi-
librium predictions and without regard to dissolution rates. For short contact times
between the water and the pipes, as will occur as water flows through the pipes, the
rates may become important in controlling dissolved lead concentrations.
• When orthophosphate was present, the release of lead to solution was slower because
the precipitation of a lead phosphate phase sequestered lead that would otherwise
have been released to solution from the hydrocerussite.

Executive Summary | xxv
RECOMMENDATIONS TO WATER UTILITIES
Information on the specific water chemistry and corrosion products is valuable in predict-
ing and controlling lead release from scales on lead service lines. When collecting samples for
compliance with the Lead and Copper Rule, utilities could gain insights into processes controlling
lead concentrations by measuring pH, dissolved inorganic carbon concentrations or alkalinity, free
and/or combined chlorine concentrations, and orthophosphate concentrations. Information on the
identity of the corrosion products can be gained by sampling and analyzing portions of pipe scales
when lead service lines are removed from a system. Archiving of sections of these lead service
lines may also be useful for experimentally testing the response of lead concentrations to proposed
changes in distribution system water chemistry.
The effectiveness of corrosion control strategies will vary depending on the source water
chemistry and the composition of the pipe scales. For pipe scales with significant amounts of the
lead(IV) oxides plattnerite or scrutinyite, low lead concentrations are achieved when a free chlorine
residual is maintained. In the absence of free chlorine, the lead(IV) oxides will breakdown, but the
release of lead to solution can be mitigated by control of the water chemistry. Less lead is released
at higher pH values. The dissolution of hydrocerussite, which is a frequently observed corrosion
product, was also slower at high pH. While increasing alkalinity can help in achieving and stabi-
lizing higher pH values, associated increases in dissolved inorganic carbon can actually increase
rates of lead release from plattnerite and had complex effects of hydrocerussite dissolution. The
addition of orthophosphate dramatically decreased rates of lead release from both plattnerite and
hydrocerussite. One mechanism through which orthophosphate mitigates lead release is the forma-
tion of hydroxylpyromorphite. Hydroxylpyromorphite has a low solubility and dissolves slowly.
Low lead concentrations can be maintained with 1 mg P/L orthophosphate and possibly less. A
minimal dose may be established that mitigates lead release effectively and minimizes cost and the
potential for negative side effects of biological growth and the need for orthophosphate removal
in wastewater treatment.

CHAPTER 1
INTRODUCTION
SIGNIFICANCE OF LEAD CORROSION TO DRINKING WATER QUALITY
Lead release from lead service lines is a serious threat to public health. The recent observa-
tions of extremely high lead levels in Washington D.C. tap water illustrate the current importance
of this problem (Edwards and Dudi 2004; Renner 2004; U.S. Environmental Protection Agency
2007). While advances in water treatment, supply, and distribution were ranked fourth on the list
of 20th Century engineering achievements compiled by the National Academy of Engineering,
management of our water treatment and supply infrastructure has been identified as a great engi-
neering challenge for the 21st century, and corrosion control plays a leading role in this challenge
(Edwards 2004). The American Society of Civil Engineers gave drinking water a D- in its 2004
report card for infrastructure (American Society of Civil Engineers 2005). As the water supply
community meets the challenge of an aging infrastructure, there is an increasing need for research
to provide a scientific basis for corrosion control.
the lead service lines and other components of premise plumbing. Lead release can occur within
components owned by the utility or in the plumbing of individual connections. The release of lead
from metal pipes and fittings is controlled by both the water composition and the properties of
the pipe surface. Lead may be released directly from the pipe or from lead-containing corrosion
products on the pipe surface. Lead corrosion products observed in lead pipes include lead(II) car-
bonates and phosphates (Schock 1999) and, more recently, lead(IV) oxide (PbO2(s)) (Edwards and
Dudi 2004; Lytle and Schock 2005). Figure 1.1 conceptually illustrates the formation of corrosion
products in scales on lead pipes and the reactions between the corrosion products and the water in
the pipe section. Solubility and dissolution rates of corrosion products are affected by water chem-
istry parameters including pH, dissolved inorganic carbon, orthophosphate, and the concentration
and type of disinfectant residual. While several studies have evaluated the equilibrium solubility of
lead corrosion products, there have been few studies of dissolution and transformation rates. Water
quality management will benefit from a systematic investigation of the effects of water chemistry
on the dissolution rates of lead corrosion products.
Information on lead corrosion rates is particularly valuable as water suppliers consider
process changes that affect water chemistry such as switching disinfectant type or dose, adjusting
pH, or adding a corrosion inhibitor. While the formation of lead corrosion products can passivate
the pipe surface and limit further lead release, changes in water chemistry can cause dissolution of
the corrosion products and lead release to solution (AWWA 2005). The increase in lead concentra-
tions in the Washington D.C. Water and Sewer Authority service area occurred following a switch
to chloramination for disinfection, which dramatically demonstrates the lead release that can result
from changes in water chemistry. An improved understanding of the effects of water chemistry
on lead release rates can be used to identify strategies that inhibit dissolution and minimize lead
release. Information on corrosion product dissolution rates can also help quantify the influence of
stagnation time on lead concentrations to which consumers are exposed and upon which monitor-
ing protocols are based.

2 | Water Chemistry Effects on Dissolution Rates of Lead Corrosion Products
Figure 1.1 Formation and stability of lead-containing corrosion products in the scales of
lead pipes
PROJECT OBJECTIVES AND OVERVIEW OF RESEARCH APPROACH
Project Objectives
The overarching project objective was to provide new information to the water supply
community that advances understanding of lead corrosion product dissolution and transformation
rates. Three specific objectives and their significance are described below.
Objective 1: Summarize the state-of-knowledge regarding dissolution rates of lead corrosion

products.
Information about the dissolution rates of lead-containing solids is distributed in the litera-
ture of fields including corrosion science and geochemistry as well as water treatment and supply.
Compiling and synthesizing information from the literature will make the results and implications
of those previous studies more accessible to the water supply community.
Objective 2: Determine dissolution and transformation rates of lead corrosion products as a

function of pH, orthophosphate concentration, dissolved inorganic carbon, and chloramines.
Currently available information on the dissolution rates of lead corrosion products is lim-
ited and primarily case-specific. Development of rate equations for lead carbonate, phosphate, and
oxide dissolution will create a chemical reaction-based framework that facilitates the prediction of
lead release rates in response to changes in water chemistry.
Objective 3: Evaluate lead release rates From corrosion products on metal pipes by applying
dissolution rate information developed for Objective 2.
To be relevant to actual systems, the dissolution rates determined using pure lead-containing
solids must be applicable to those same lead-containing solids when they are present as corrosion
products on actual pipe.

Chapter 1: Introduction | 3
Figure 1.2 Overview of research approach
Overview of Research Approach and Structure of the Report
The research approach was organized into three integrated tasks to pursue the three objec-
tives (Figure 1.2). Task 1 was completion of a literature review regarding the dissolution rates of
lead(II) carbonate, lead(II) phosphate, and lead(IV) oxide precipitates. Task 2 was a systematic
experimental investigation of the dissolution rates of three important lead corrosion products:
hydrocerussite, hydroxypyromorphite, and lead(IV) oxide. Dissolution rates were measured as
functions of the important water chemistry parameters pH, dissolved inorganic carbon concentra-
tion, dissolved orthophosphate concentration, and monochloramine concentration. Task 3 extended
the study of dissolution rates of pure solids to release rates from pipe scales. The results of Task 1
are presented in Chapter 2 and Appendix A of this report. Chapter 3 presents the results of the dis-
solution rate measurements, and Chapter 4 presents the results of the experiments with scales on
real lead pipes. The findings from all three tasks are summarized in Chapter 5, and Chapter 6 then
presents recommendations for utilities.
BACKGROUND ON LEAD CORROSION AND CORROSION CONTROL
Lead Corrosion
The internal corrosion of lead-containing pipe, fittings, and solder in premise plumbing is
currently the most significant source of lead to drinking water. Lead can be released from pure lead
materials as well as from brass, which can contain significant lead amounts depending on the alloy
(Dudi et al. 2005). Although new construction does not use lead pipe, many older buildings retain
original lead service lines and internal plumbing. Concern over lead concentrations in tap water
motivated the passage of the Lead and Copper Rule (LCR) in 1991. The LCR set an action level
of 0.015 mg/L for lead. Utilities must implement methods to control lead corrosion if the 90th per-
centile of samples exceeds the action level (U.S.EPA 1991). Lead concentrations in drinking water
are strongly influenced by distribution system water chemistry. The highest concentrations are
encountered in systems with water of relatively low pH and low alkalinity (Dodrill and Edwards
1995; Schock 1999). In response to changes in water chemistry, high lead concentrations can also
be observed in systems with no previous history of a lead problem. The increase in lead concentra-
tions in the Washington D.C. Water and Sewer Authority service area that occurred following a
switch to chloramination for disinfection is a dramatic example of lead release that can result from

changes in water chemistry (Edwards and Dudi 2004; Renner 2004). This change in disinfectant
was initiated to meet the regulatory requirements for disinfection byproducts. However, the lower
oxidation-reduction potential with chloramines than with free chlorine resulted in the break down
of lead(IV) oxide phases and the release of lead to solution (Lytle and Schock 2005; Vasquez et
al. 2006).
Corrosion Products
Lead corrosion products observed on pipe surfaces include lead(II) carbonates and phos-
phates (Schock 1999) and, more recently, lead(IV) oxides (PbO2(s)) (Edwards and Dudi 2004; Lytle
and Schock 2005). The conditions under which these corrosion products are formed are discussed
below. More detailed innformation on the solubility of these phases is then included in Chapter 2.
The lead carbonates cerussite (PbCO3(s)) and hydrocerussite (Pb3(CO3)2(OH)2(s)) are
formed in systems for which the solution pH and dissolved inorganic carbon are sufficiently high
to be saturated with respect to these solids. Previous studies have directly observed cerussite and
hydrocerussite on lead and brass materials using X-ray diffraction (XRD) (Hozalski et al. 2005),
cyclic voltammetry and electrochemical impedance spectroscopy (El-Egamy 1995), and extended
X-ray absorption fine structure spectroscopy (Frenkel and Korshin 1999). In many cases the cor-
rosion layers are similar to pure phases but are more disordered (Frenkel and Korshin 1999).
Cerussite and hydrocerussite can also form at the surfaces of calcium carbonate solids through a
reaction sequence involving lead adsorption, calcium carbonate dissolution, and lead carbonate
re-precipitation (Godelitsas et al. 2003; Rouff et al. 2004).
Lead phosphate solids are only likely to form in systems to which orthophosphate has been
added as a corrosion inhibitor. Hydroxypyromorphite (Pb5(PO4)3OH(s)) is a commonly observed
lead phosphate corrosion product. For example, a recent study of pipe in Glasgow, Scotland used
XRD and secondary ion mass spectrometry (SIMS) to identify lead phosphate solids as corrosion
products (Davidson et al. 2004). In the presence of chloride, chloropyromorphite (Pb5(PO4)3Cl(s))
can form as a very insoluble lead corrosion product. Chloropyromorphite was recently observed
in corrosion products in lead pipes, but the dissolved lead concentrations were still higher than the
equilibrium solubility of chloropyromorphite (Schock et al. 2005a).
Lead(IV) oxides have recently been observed as a component of the scales on lead pipe
from several communities (Lytle and Schock 2005). Two polymorphs of PbO2 can form: scrutinyite
(α-PbO2) and plattnerite (β-PbO2) (Lytle and Schock 2005). Although lead is not stable in the +IV
oxidation state in oxygenated water, this oxidation state is stable at the high oxidation-reduction
potential (ORP) provided by free chlorine. Lead(IV) precipitates in low solubility PbO2(s) phases
that can maintain low dissolved lead concentrations. Because PbO2 is only stable at high ORP,
the decrease in the ORP that occurs when chloramines are used instead of free chlorine causes the
PbO2 to breakdown and release lead back to solution (Vasquez et al. 2006). Laboratory and pilot-
scale studies have observed higher dissolved lead levels with chloramines than with free chlorine,
which can be explained by the differences in solubility of Pb(IV) and Pb(II) phases (Edwards and
Dudi 2004; Vasquez et al. 2006). The formation of PbO2 was observed upon reaction of dissolved
lead with sodium hypochlorite. The formation of PbO2 is reversible, with PbO2 disappearing when
a residual hypochlorite concentration is not maintained.

Chapter 1: Introduction | 5
Corrosion Control
Many water utilities must implement a corrosion control strategy to minimize dissolved
lead concentrations in tap water. Recommended strategies for controlling lead corrosion include
pH and/or alkalinity adjustment and addition of orthophosphate (Edwards et al. 1999; Schock
1999). The general strategy of corrosion control is to promote the formation of low solubility lead-
containing solids that passivate the surface.
Based on a survey of utilities, large decreases in soluble lead levels are observed with
increasing alkalinity. The largest decreases occur as the alkalinity increases from less than 30 mg/L
to greater than 30 mg/L as CaCO3 (Edwards et al. 1999). The effect of increasing alkalinity is
consistent with lead concentrations being determined by the solubility of lead carbonate (cerussite
and hydrocerussite) solids. The stability of these carbonate solids is determined by the activity
of the carbonate ion, which is determined by both the DIC (or alternatively alkalinity) and pH.
Consequently, increasing pH without also increasing carbonate alkalinity will not be as effective.
In fact, without sufficient alkalinity (less than 20–30 mg/L as CaCO3), increasing pH can actu-
ally exacerbate lead release (Dodrill and Edwards 1995). In some systems, lead decreases with
increasing alkalinity may also be related to the formation of a protective calcium carbonate layer.
New information from this project can help water suppliers evaluate alternatives of adjusting pH,
dissolved inorganic carbon, or both.
Phosphate inhibitors can effectively mitigate lead concentrations in drinking water (Edwards
et al. 1999; Schock 1999; Churchill et al. 2000; Cantor et al. 2003). The form in which the phos-
phate is added is vitally important. Orthophosphate and associated forms (e.g., zinc orthophos-
phate) are the most effective inhibitors. Orthophosphate inhibitors can result in the precipitation
of insoluble lead phosphate solids that limit lead release to solution. In contrast, polyphosphates
form dissolved complexes with lead and can increase dissolved lead concentrations in tap water
(Dodrill and Edwards 1995; Edwards and McNeill 2002; McNeill and Edwards 2002; Schock et
al. 2005a; Cantor 2006). The benefits of orthophosphate as a corrosion inhibitor are most signifi-
cant for systems with relatively low alkalinity or pH, conditions at which lead concentrations are
not controlled by lead carbonate solids. In a survey of utilities, the most pronounced benefits of
orthophosphate addition were seen in waters with less than 30 mg/L alkalinity as CaCO3 (Dodrill
and Edwards 1995). Recommended orthophosphate doses can range from less than 1 mg P/L to
as much as 4.5 mg P/L (Edwards et al. 1999). The results of this project can be used to evaluate
alternatives of either adjusting the pH or adding orthophosphate for systems with neutral pH and
high dissolved lead concentrations.
Although orthophosphate addition can effectively control lead concentrations, it can have
detrimental effects such as the promotion of bacterial growth within the distribution system and
premise plumbing. A recent pipe loop study using orthophosphate observed a significant increase
in bacteria as measured by heterotrophic plate counts (Hozalski et al. 2005). A survey of 2500 utili-
ties that had exceeded lead or copper action limits indicated that most utilities are aware of the
potential drawbacks of orthophosphate addition; however, utilities still select inhibitors based on
only limited information (McNeill and Edwards 2002). The proposed project will generate new
information on the rates of lead phosphate dissolution that can be beneficial to utilities designing
a lead control strategy.
Stannous chloride (SnCl2) is another chemical that can be used to inhibit lead corrosion. A
recent study showed that 0.125 mg/L stannous chloride addition was almost as effective as 1 mg/L
P orthophosphate addition in limiting lead concentrations, and the system with stannous chloride

addition had substantially lower microbiological growth (Hozalski et al. 2005). Recent investiga-
tions have also demonstrated that aeration can be an effective lead corrosion control strategy for
systems that have sufficient dissolved inorganic carbon. Aeration increases the pH by removing
carbon dioxide; however, aeration can increase dissolved oxygen and consequently the oxidation
potential that causes lead corrosion. Aeration may also introduce microbiological contaminants
(Schock 1989; Lytle et al. 1998; Schock et al. 2002).
Lead Release Rates From Pipe
Estimates of lead concentrations in tap water based on equilibrium solubility are qualita-
tively good, but they are usually not quantitatively accurate. Equilibrium-based models tend to
overpredict lead concentrations. Such models are limited by the accuracy of availabile equilibrium
constants, transitions between scale types, by-product release, and reaction kinetics (Edwards et al.
1999; Vasquez et al. 2006). For systems that are not at equilibrium, measurements of dissolution
rates can improve quantitative estimates of soluble lead concentrations in lead pipes and down-
stream tap water.
Differences between lead concentrations during times of flowing water and during stag-
nation periods can be significant. For example, lead concentrations in new pipe were low during
flowing conditions but exceeded 15 μ��
��
g/L after 8 hours of stagnation, even in the presence of inhibi-
tors (Hozalski et al. 2005). The role of stagnation time reflects the role of dissolution rates. During
stagnant periods, dissolved lead concentrations in lead pipe approached equilibrium exponentially
with time. The largest increases occur within the first 24 hours. Release rate data could be fit using
a radial diffusion model with a diffusion barrier term, but this model may not be mechanistically
accurate (Lytle and Schock 2000). The following chapter includes information on the equilibrium
solubility of lead corrosion products and introduces an approach to examining dissolution rates.

CHAPTER 2
REVIEW OF LEAD CARBONATE, PHOSPHATE, AND OXIDE
DISSOLUTION RATES AND SOLUBILITY
APPROACH TO THE REVIEW
A review of the current state-of-knowledge regarding dissolution rates of lead carbonate

and phosphate solids was written. While there are several excellent reviews of corrosion processes,
corrosion control, and the influence of equilibrium solubility on corrosion products (Schock 1999;
MWH 2005), there is no review dedicated to the dissolution rates of these corrosion products.
A literature search on dissolution rates of lead carbonate, phosphate, and oxide solids was con-
ducted and the references found in the search were critically reviewed. The review examined refer-
ences from fields including water supply and treatment, corrosion science, aquatic chemistry, and
geochemistry.
SUMMARY OF THE REVIEW
The full review is included as Appendix A of this report. The review introduces the impact
of the dissolution of lead corrosion products on drinking water quality. It then presented an over-
view of the equilibrium solubility of lead(II) carbonates and phosphates and lead(IV) oxides and
the effects of pH and other water chemistry parameters on solubility. The dissolution rates of
lead(II) carbonate and phosphate solids are then organized. The presentation of rates describes an
approach that accounts for the effects of pH and of the distance of the solution from equilibrium
solubility. Due to the limited studies of lead-containing solids, general observations are also drawn
from previous work on the dissolution rates of other carbonate and phosphate solids that have
similar structures to those of hydrocerussite and hydroxylpyromorphite. For the sake of providing
background information, the rest of this chapter includes the most essential information regarding
the solubility and dissolution rates of the lead-containing solids.
LEAD SOLUBILITY
Maximum dissolved lead concentrations are determined by the equilibrium solubility of

the corrosion products. Equilibrium solubility has been thoroughly reviewed by Schock (Schock
1999), and only a brief overview is provided here. The solubility of hydrocerussite and hydroxy-
pyromorphite are controlled by their equilibrium solubility products (Ksp) given as equations 2�1,
2.3, and 2.5 for the dissolution reactions 2.2, 2.4, and 2.6, respectively.
[Pb 2 +] 3 [CO 23 −] 2
Ksp, hydrocerussite = = 10 − 18.77 (2.1)
[H +] 2
Pb3(CO3)2(OH)2(s) + 2H+ = 3Pb2+ + 2CO32– + 2H2O (2.2)

(a) (b)
-2 -2
-3 -3
hydrocerussite
-4 -4
cerussite Pb(OH) 2(s)
Dissolved Pb (M)
Dissolved Pb (M)
-5 -5 hydroxypyromorphite Pb(OH) 2(s)
-6 -6
-7 -7
-8 -8
dissolved Pb
dissolved Pb
-9 -9
4 5 6 7 8 9 10 11 4 5 6 7 8 9 10 11
pH pH
Figure 2.1 Lead solubility as a function of pH. Regions are shown for dominant dissolved
lead and precipitated solid phases for (a) 30 mgC/L DIC (2.5·10–3 M) and (b) 3 mgC/L DIC
(2.5·10–4 M) plus 3 mg/L dissolved orthophosphate (9.7·10–5 M). The presence of orthophos-
phate significantly lowers the dissolved lead concentrations at neutral to low pH.
[Pb 2 +] 4 [PO 34 −] 3
Ksp, hydroxypyromorphite = = 10 − 62.79 (2.3)
[H ] +
Pb5(PO4)3OH(s) + H+ = 5Pb2+ + 3PO43– + H2O (2.4)
[Pb 4 +]
Ksp, scrutinyite = = 10 − 8.26 (2.5)
[H +] 4
PbO2(s) + 4H+ = Pb4+ + 4H2O (2.6)
The pH, as expressed by [H+], carbonate ion concentrate ion [CO32–], and orthophosphate
concentration [PO43–] determine the equilibrium concentration of the free lead cation [Pb2+] or
[Pb4+]. The total dissolved lead concentration is then the sum of all dissolved lead species, which
includes the free ion as well as hydrolysis complexes (e.g., PbOH+), complexes with inorganic
anions (e.g., PbCO3(aq)), and complexes with organic ligands (e.g., humic substances in natural
organic matter). For example, natural organic matter increases lead solubility and inhibits forma-
tion of hyrocerussite and cerussite (Korshin et al. 2000; Korshin et al. 2005). Higher dissolved lead
concentrations have also been observed in the presence of chloramines, which may result from the
formation of soluble lead complexes but is more likely due to changes in the ORP that result in the
dissolution of lead(IV) oxides (Vasquez et al. 2006). A complete set of reactions used in equilib-
rium calculations in this work is included in Appendix B.

Chapter 2: Review of Lead Carbonate, Phosphate, and Oxide Dissolution Rates and Solubility | 9
(a) (b)
1.6 1.6
PbO2(s) PbO2(s)
1.2 1.2
PbO32-
PbO32-
0.8 0.8
Pb5(PO4) 3 OH(s)
Pb2+ Pb2+
0.4 0.4
PbCO3(aq)
EH (V)
EH (V)
Pb(OH)42-
Pb(OH)42-
0.0 PbHCO3+ Pb(OH)2(s) 0.0 Pb(OH)2(s)
-0.4 -0.4
-0.8 Pb(s) -0.8 Pb(s)
-1.2 -1.2
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
pH pH
Figure 2.2 Predominance area diagrams showing the dominant lead solid phase of dis-
solved species as a function pH and oxidation-reduction potential. Diagrams are shown for
(a) 30 mgC/L DIC (2.5·10–3 M) and (b) 3 mgC/L DIC (2.5·10–4 M) plus 3 mg/L dissolved ortho-
phosphate (9.7·10–5 M). The diagrams are constructed for a total lead concentration of 15 μg/L.
The dashed lines represent the stability limits of water (PO2 = 0.21 atm and PH2 = 1 atm).
Both the carbonate and orthophosphate ion concentrations are strongly pH-dependent. The
carbonate ion concentration can be determined from the pH and dissolved inorganic carbon con-
centration (DIC) (2.7).
DIC = [H2CO3*] + [HCO3–] + [CO32–] (2.7)
Many studies have presented results based on alkalinity instead of DIC. If carbonate species domi-
nate the alkalinity (2.8), then DIC can be calculated from pH and alkalinity.
Carbonate Alkalinity = –[H+] + [OH–] + [HCO3–] + 2[CO32–] (2.8)
However, as Schock makes clear in his review, the carbonate ion concentration is the property that
truly determines lead carbonate solubility, and that concentration is most appropriately determined
by the relationship between pH and DIC. The phosphate ion concentration can similarly be deter-
mined by the pH and total dissolved orthophosphate concentration.
The expected solid phase, as well as the presence of any solid phase, at equilibrium depends
upon the composition of the aqueous phase. In the absence of orthophosphate, cerussite and hydro-
cerussite will form, with hydrocerussite favored at higher pH (Figure 2.1a). In the presence of
both DIC and orthophosphate, hydroxypyromorphite becomes the expected phase as DIC and/
or pH decrease (Figure 2.2b). At sufficiently high pH, lead hydroxide (Pb(OH)2(s)) precipitation
is predicted. The lead(II) phases just discussed are formed at oxidation-reduction potentials of

aerated solutions and solutions treated with chloramines. At the higher ORP values established
when hypochlorite is present, lead(IV) oxide forms as the most stable phase. Figure 2.2 illustrates
the favored solid and dissolved species as a function of both pH and ORP. Changes in ORP that
result from changes in the residual disinfectant or in disinfectant demand in the lead pipe can lead
to transformations from low solubility PbO2(s) to higher solubility lead(II) carbonate and phosphate
phases.
Dissolution Rates of Precipitated Solids
Dissolution Rate Law
Dissolution rates of precipitated solids are functions of properties of both the dissolving
solid and the solution in which it is dissolving. A general dissolution rate law (Eq. 2.9) has been
developed based on chemical thermodynamics (Lasaga 1998).
Rate = k0 {H +} nH+ {i} ni ]1 − 10 SI g

Ω
(2.9)
[Pb 2 +] 3 [CO 23 −] 2 [H +] − 2
SI = log d IAP n = log e o (2.10)
Ksp Ksp, hydrocerussite
The rate is expressed in units of mol m–2 s–1 and is applicable for a given temperature. The
dissolution rate constant k0 is also in mol m–2 s–1. The effects of pH are included by considering
the proton activity {H+} and the order of the reaction with respect to H+ (nH+). Additional dissolu-
tion enhancing or inhibiting species are expressed as {i} with ni as their reaction orders; possible
inhibiting or enhancing species include orthophosphate, polyphosphates, carbonate, and complex-
ing ligands. The last term includes the saturation index (SI), which expresses the ratio of the ion
activity product of a solid to its equilibrium solubility product, shown in Equation 2.10 for the case
of hydrocerussite. The SI term accounts for the extent of saturation (i.e., supersatured vs. under-
satured) of the solution with respect to the dissolving solid. The term Ω is an empirical coefficient
to account for the non-elementary nature of dissolution reactions; values are usually in the range
of 0.5 to 1.0. The greater the extent of undersaturation (i.e., SI is more negative), the higher the
dissolution rate. As the solution approaches equilibrium with respect to the solubility of the dis-
solving solid, SI approaches zero and the net dissolution rate approaches zero. For supersaturated
solutions, SI is positive, and the dissolution rate becomes negative (i.e., it becomes a precipitation
rate). Note that pH, orthophosphate, and carbonate can affect the dissolution rate by two possible
mechanisms: (1) directly as inhibiting or enhancing species and (2) by altering the solution satura-
tion state. The rate law was originally developed for the interpretation of silicate mineral dissolu-
tion rates (Lasaga 1995; Lasaga 1998), but it has subsequently been applied to the dissolution of
oxides, carbonates, and phosphates (Giammar and Hering 2000; Giammar 2001; Giammar and
Hering 2002).

Chapter 2: Review of Lead Carbonate, Phosphate, and Oxide Dissolution Rates and Solubility | 11
Dissolution Rates of Lead Carbonates, Phosphates, and Oxides
Several previous studies have observed carbonate dissolution rates that follow Equation 2.9.
An overview of divalent metal carbonate dissolution determined that rates increased in the sequence
Ni < Mg < Co < Fe < Mn < Zn < Cd <Sr < Ca ≈ Ba ≈ Pb, and dissolution rates of PbCO3(s) were
on the order of 10–7 mol m–2 s–1 (Pokrovsky and Schott 2002). The rates fit Equation 2.9 by con-
sidering the term for solution saturation and two enhancing species, which were identified as the
surface species that are most vulnerable to dissolution. Pokrovsky and co-workers have further
developed the rate dependence on saturation and surface speciation for the specific case of magne-
site (MgCO3(s)) (Pokrovsky and Schott 1999; Pokrovsky et al. 1999; Oelkers et al. 2002), and they
have successfully modeled both dissolution and precipitation with the same rate equation. Studies
of calcite dissolution (Gledhill and Morse 2004; Cubillas et al. 2005) and precipitation (Ferris et
al. 2004) have observed a similar dependence of rates on solution saturation.
The equilibrium solubility and precipitation rates of lead phosphate minerals have been
investigated in previous studies that were motivated by an interest in precipitating lead phosphate
minerals for in situ immobilization of lead in soil or groundwater (Nriagu 1973; Ma et al. 1995;
Hettiarachchi et al. 2000; Yang et al. 2001; Cao et al. 2002; Ryan et al. 2004). Chloropyromorphite
(Pb5(PO4)3Cl(s)), a phase closely related to hydroxypyromorphite, has received the most attention.
Precipitation of chloropyromorphite occurs rapidly and is usually homogeneous (i.e., solids nucle-
ate directly from solution without needing a pre-existing surface) (Lower et al. 1998; Manecki et
al. 2000). In contrast to precipitation, the dissolution rates of lead phosphates have not received
much attention. Current work in our laboratory on the dissolution rates of chloropyromorphite has
observed that dissolution rates increase with decreasing pH, decreasing solution saturation, and
decreasing crystallinity of the solid phase (Xie and Giammar 2007).
Only recently have dissolution rate measurements of lead(IV) oxides been made. Because
the reductive dissolution of PbO2 in pure water is energetically favorable, extensive dissolution
was measured for plattnerite (Lin and Valentine 2008a) and scrutinyite (Dryer and Korshin 2007).
The dissolution of both materials was significantly accelerated by the presence of natural organic
matter (NOM), which served as a reductant for the PbO2. NOM may also inhibit the development
of Pb(II) precipitates that would limit further dissolution. The dissolution of PbO2 increased with
decreasing pH. At pH 7, the data of Lin and Valentine (2008a) can be used to extract a dissolu-
tion rate of 1.8·10–12 mol m–2 s–1. This study also verified that the lead released was lead(II) and
not lead(IV). Recent research found that iodide could be oxidized to iodate, which would occur
through coupling with reductive dissolution of PbO2, and the rates were determined to be functions
of the PbO2 surface area, iodide concentration, and pH (Lin et al. 2008).

CHAPTER 3
DISSOLUTION RATES OF LEAD CORROSION PRODUCTS
OVERVIEW OF RESEARCH ON DISSOLUTION RATES
Dissolution rates of three important lead corrosion products, the lead carbonate hydro-
cerussite, the lead phosphate hydroxypyromorphite, and the lead(IV) oxide plattnerite, were quan-
titatively determined in completely-mixed continuous-flow reactors (Table 3.1). The dissolution
rates were determined as a function of pH, orthophosphate concentration, dissolved inorganic
carbon concentration, and the presence of chloramines. The emphasis of the research is on deter-
mination of dissolution rates as opposed to equilibrium solubility of the three solid phases; the
ability of equilibrium calculations versus rate equations to predict dissolved lead concentrations
is discussed for each solid. In addition to monitoring lead release to quantify dissolution rates, the
solid phases in selected experiments were characterized to evaluate possible transformation of the
solid phases. Transformations were expected if the solution chemistry promoted the dissolution of
one phase and the precipitation of another. The solution composition was evaluated in a reaction-
based framework, and the dissolution rates were organized to determine the degree to which a
general predictive model could be developed.
MATERIALS AND METHODS
Preparation and Characterization of Lead Corrosion Products
Hydrocerussite
Hydrocerussite (Pb3(CO3)2OH(S)) was synthesized by precipitation of a supersaturated

solution containing carbonate and lead. The synthesis was conducted at the ambient laboratory
temperature (20°C). A 0.02 M dissolved inorganic carbon solution was prepared by dissolving
sodium bicarbonate (NaHCO3) to ultrapure water, and a 0.03 M lead nitrate (Pb(NO3)2) solution
was also prepared with ultrapure water. The dissolved inorganic carbon and lead nitrate solutions
were simultaneously added to 500 mL of ultrapure water using separate burets to maintain the
correct stoichiometric Pb:C ratio of 3:2. Aliquots of a 1.0 M sodium hydroxide solution (NaOH)
were periodically added to maintain pH 8.6 ± 0.2 throughout the synthesis process. Maintaining a
constant pH is critical in the preparation of hydrocerussite; if the pH is not maintained within the
appropriate range, then competing solids will precipitate The solid suspension was washed with
ultrapure water and dialyzed for 24 hours. The suspension was then freeze-dried to generate unal-
tered dry solids.
A sample of hydrocerussite was prepared for phase identification by X-ray powder diffrac-
tion (XRD) (Figure 3.1). The XRD pattern confirms that hydrocerussite was synthesized and that
cerrusite (PbCO3), a competing solid, did not form. Careful control of pH was required to promote
formation of hydrocerussite and not cerussite. Scanning electron microscopy shows that long rod-
like crystals formed during the precipitation of hydrocerussite (Figure 3.2) that are distinctly dif-
ferent from the crystals formed in the precipitation of cerussite (Figure 3.3).
13

Table 3.1
Solid phases studied in dissolution rate experiments
Specific
Solid surface area Importance to study
Hydrocerussite: Pb3(CO3)2(OH)2(s) 4.8 m2/g Frequently observed in scales on lead pipes
from distribution systems.
Hydroxypyromorphite: Pb5(PO4)3OH(s) 16 m2/g Formation is likely when orthophosphate is
added for corrosion control.
Plattnerite: PbO2(s) 3.6 m2/g Formed in some systems with high free
chlorine residual and is not stable with
chloramines residual.
Relative Intensity
Hydrocerussite, syn PDF#00-013-0131
20 25 30 35 40 45 50 55 60
2θ (°)
Figure 3.1 XRD pattern of hydrocerussite and reference pattern (PDF#00-013-0131)
As determined by BET-N2 adsorption, the specific surface area (m2/g) of the synthetic
hydrocerussite was 4.8 m2/g. The specific surface area is used to normalize the dissolution rates of
hydrocerussite.
Hydroxylpyromorphite
Hydroxylpyromorphite was synthesized by simultaneous titration of 250 mL of a 0.25 M

lead nitrate solution with 250 mL of a 0.15 M phosphate solution into 500 mL of ultrapure water
using a peristaltic pump. The pH was maintained at 10.0 ± 0.3 by addition of 1.0 M potassium
hydroxide solution for 24 hours and was open to the atmosphere. The solid was washed with
ultrapure water and dialyzed for 24 hours before it was freeze-dried. A portion of the synthesized
material was acid washed using a pH 3 solution of nitric acid (1 mM concentration) with 1 mM
sodium nitrate. In the acid-washing process, the suspension of solid in the pH 3 solution was stirred
overnight and then centrifuged for 5 minutes at 10,000 rpm. The supernatant was poured off and
collected. The solids were then washed with ultrapure water. This process was repeated 5 times.

Chapter 3: Dissolution Rates of Lead Corrosion Products | 15

Figure 3.2 Synthesized hydrocerussite at two different levels of magnification
Figure 3.3 Synthesized cerussite
The X-ray diffraction pattern of the hydroxylpyromorphite indicates that the correct phase
was synthesized (Figure 3.4). A comparison of the XRD patterns of the as-synthesized and acid-
washed materials shows that the materials are nearly identical. The only minor changes are the dis-
appearance of a peak at 33.8° and a decrease in the intensity of the peak at 18° after acid-washing.
The loss of the peak at 33.8° indicates that some trace phase was removed by the acid-washing pro-
cess. This peak and others are is similar but not identical to ones for hydrocerussite. It is possible
that some lead carbonate phase similar to hydrocerussite was removed by the acid-washing step.
The specific surface area of the acid-washed hydroxylpyromorphite was determined to
be 16 m2/g. The solid consisted of small particles and aggregates of particles smaller than about
100 nm (Figure 3.5).
Total digestion of the solid provided additional information regarding the composition of
the material. The stoichiometric ratio of Pb:P in pure hydroxylpyromorphite would be 1.67. Prior
to acid washing, the measured Pb:P ratio for the solid was 1.75, which is statistically significantly
higher than the ratio in the pure solid. The higher ratio suggests the potential presence of lead in a
non-phosphate form such as a lead hydroxide or lead carbonate. After acid washing and digestion,
the ratio decreased to 1.55, which is also statistically lower than expected for pure hydroxylpyro-
morphite and may indicate the presence of another lead phosphate mineral such as Pb3(PO4)2 in
addition to or in place of hydroxylpyromorphite.
The Fourier transform infrared (FTIR) spectra of the acid-washed and non-acid washed
material contained peaks that correspond to phosphate at 1000 cm–1. Small broad peaks seen in
the non-acid washed material may indicate the presence of carbonate and/or hydroxide, which are

Figure 3.4 X-ray diffraction patterns of as-synthesized and acid-washed hydroxylpyromorphite.

Patterns are shown with reference patterns for two lead phosphate solids and hydrocerussite.
Figure 3.5 Acid-washed hydroxylpyromorhophite
found in the range of 1360–1030 cm–1 and 3650–2500 cm–1, respectively (Smith,1996). Because
the peak corresponding to carbonate was not observed for the acid-washed material, it is likely
that acid-washing removed a trace amount of hydrocerussite. Raman spectra of the acid-washed
material showed all three peaks associated with phosphate and a peak associated with hydroxide,
which indicated that the material was hydroxylpyromorphite (Pb5(PO4)3OH) as the lead phosphate
solid Pb3(PO4)2 does not contain a hydroxyl group.
Plattnerite
The lead(IV) oxide plattnerite (β-PbO2) was purchased from Fisher Scientific in the form of
black particles with primary particle sizes ranging from 50 nm to 500 nm as determined by electron

Figure 3.6 Scanning electron micrograph of plattnerite

Relative Intensity
20 25 30 35 40 45 50 55 60
2θ (°)
Figure 3.7 X-ray diffraction patterns of plattnerite. Patterns are included for the purchased
material (top) and the reference pattern (PDF# 01-071-4820).
microscopy (Figure 3.6). The solid was indentified as pure plattnerite by XRD (Figure 3.7). The
measured specific surface area of the plattnerite was 3.6 m2/g.
Efforts were also made to synthesize scrutinyite (α-PbO2), which is the other common
polymorph of lead (IV) oxide. Syntheses involved the oxidation of dissolved lead(II) or a lead(II)-
containing solid by an excess of free chlorine. The resultant product often was a mixture of
plattnerite and scrutinyite. The distribution of the products between plattnerite and scrutinyite
was a function of the pH, the initial lead concentration, the dissolved inorganic carbon concen-
tration, and the lead(II) precursor. Because scrutinyite could only be synthesized in a pure form
when starting with relatively low total concentrations of lead, only limited amounts of this phase
could be formed.

Characterization Methods for Lead Corrosion Products
The three solid phases were characterized by X-ray powder diffraction (XRD), scanning
electron microscopy (SEM), Fourier transform infrared (FTIR) and Raman spectroscopy, surface
area analysis, and elemental analysis of digested solids. Together these methods provided informa-
tion on the composition, structure, and morphology of the solid phases. Characterization was per-
formed on the pristine as-synthesized solids and also on solids retrieved following their reaction in
aqueous solutions of controlled composition.
XRD was used to identify the crystalline phases present initially and following dissolu-
tion experiments. XRD was done on a Rigaku Geigerflex D-MAX/A diffractometer using Cu-Ka
radiation. The instrument is equipped with a vertical goniometer and a scintillation counter. SEM
provided images of the lead corrosion products. SEM was performed on a Hitachi model S-4500
field emission scanning electron microscope with a NORAN Instruments energy dispersive X-ray
(EDX) microanalysis system. Changes in particle morphology and spacing can be determined
using SEM. The specific surface area (m2/g) of each lead corrosion product was determined by the
Brunauer, Emmett, and Teller (BET) N2-adsorption isotherm. The specific surface area was used
to normalize the dissolution rates of each lead corrosion product. A Nexus 470 Thermo Electron
FTIR was used with a single-reflection attenuated total reflectance germanium crystal accessory.
Raman spectroscopy was performed using a HoloLab Series 5000 laser Raman microprobe (Kaiser
Optical System, Inc.). Solids were irradiated by a 100-mW, 532-nm, frequency-doubled Nd-YAG
visible laser, which delivered about 11mW laser power to the solid surface.
For analysis of the composition and stoichiometry of the hydroxylpyromorphite, the solids
were subjected to a total digestion procedure. Solids were digested completely in an acid-digestion
bomb. Approximately one milligram of hydroxylpyromorphite was added to a bomb containing
10 mL of concentrated nitric acid. The bomb was heated at 175 °C for approximately 20 hours.
After cooling to room temperature all of the liquid was collected for dilution and analysis of dis-
solved concentrations of lead and phosphorus.
Analysis of Aqueous Composition
Dissolved concentrations of lead and phosphorus were determined by inductively coupled

plasma mass spectroscopy (ICP-MS) on an instrument that has an instrument detection limit of
9 ppt (ng/L) and a method detection limit of 50 ppt (ng/L) for lead. Analyses were performed on
an Agilent 7500ce instrument equipped with an Octopole Reaction System (ORS). The pH of
solutions was measured with a glass pH electrode and pH meter (Accumet). Free chlorine and
chloramine concentrations were measured using the standard DPD colorimetric method (Clesceri
et al. 1999). A scanning ultraviolet/visible spectrophotometer (Perkin-Elmer Lambda 2S) was used
for analysis.
Measurement of Dissolution Rates
Dissolution rates of the solids are quantified by using small stirred flow-through reactors
(Figure 3.8). The reactors have volumes of 84 mL. The reactors were loaded with 1 g/L suspensions
of solids and sealed with 0.45 μm filter membranes. The influent is supplied to the reactors with a
peristaltic pump set at a flow rate to provide a hydraulic residence time of either 30 or 60 minutes;
residence times of 60 minutes were used for hydrocerussite experiments and of 30 minutes for

0.45 µm
filter
solid
suspension Influent solution of
controlled composition
ll co ion
[Pb]out, [[P]out , pH
Figure 3.8 Flow-through reactor for measuring dissolution rates
Table 3.2
Experimental variables in systematic investigation of dissolution rates
Property Selected conditions Importance to study
pH 7.5, 8.5, 10 Range found in many systems, important variable in
susceptibility to corrosion
Dissolved inorganic 0, 10, 50 mgC/L Inhibits corrosion by formation of lead carbonate
carbon corrosion products
Orthophosphate 0, 1, 4 mg/L as P Inhibits corrosion at some conditions, formation of
insoluble lead phosphates
Chloramines 0, 2 mg/L as Cl2 Potentially enhances dissolution
hydroxylpyromorphite and plattnerite experiments. The influent compositions were controlled to

evaluate the effects of pH, orthophosphate, dissolved inorganic carbon, and chloramines on the
dissolution rates (Table 3.2). Each combination of pH and DIC corresponds to a different alkalinity
value (Appendix C), and for this study DIC was used as the independent variable because alkalin-
ity varies with pH for a given DIC. Effluent samples were collected and analyzed for dissolved
lead (Pb) by ICP-MS. Volumetric flow rate and pH were periodically measured throughout each
experiment.
The reactor influents were prepared in 5 L or 10 L tedlar bags to ensure no transfer of
carbon dioxide into or out of solution. To minimize uptake of dissolved carbon dioxide, the ultra-
pure water in each influent was boiled and/or purged with nitrogen immediately before being
pumped into the tedlar bags. The influent pH was adjusted by the addition of sodium hydroxide.
Dissolved inorganic carbon (DIC) concentrations were provided by the addition of sodium bicar-
bonate (NaHCO3). Orthophosphate concentrations are set by the addition of sodium phosphate
(NaH2PO4) from a stock solution. Chloramine stock solutions (200 mg/L as Cl2) for addition to
the influent solutions were prepared immediately before each dissolution experiment because the
monochloramine concentration decays approximately 10% each day. The stock solutions were
prepared by mixing volumes of 6% (w/w) sodium hypochlorite (NaOCl) and 2500 mg/L (as NH3)
ammonium chloride solutions in ultrapure water. The stock chloramines solution concentration
was then quantified by titration with standardized sodium thiosulfate (Na2S2O3) solution and fur-
ther confirmed by the DPD colorimetric standard method. A 0.79 Cl2:N molar ratio (4:1 Cl2:N
mass ratio) was used to best simulate the conditions in drinking water distribution systems and

Ceq
C0
[Pb]eff
Css
Time (h)
Figure 3.9 Illustration of the dissolved lead effluent from a flow-through reactor. The illus-
tration is for a hydraulic residence time of 1 hour. The initial concentration (C0), equilibrium
solubility (Ceq), and steady-state concentration (Css) are identified.
premise plumbing. Under these conditions the dominant form of chloramine is monochloramine
(NH2Cl). A 1.0 M NaNO3 solution is then injected into the tedlar bag to set the ionic strength at
0.01 M.
For experiments with hydrocerussite and hydroxylpyromorphite and without monochlora-
mine, a 0.1 M buffering solution (HEPES for pH = 7.5, TAPS for pH = 8.5, and CAPS for pH =
10) was injected to provide a buffer solution of 0.001 M. Because some of the buffers were found
to accelerate plattnerite dissolution by reducing lead(IV) to more soluble lead(II) species, no buf-
fers were used for the experiments with plattnerite. Experiments using chloramines were also not
buffered because the organic buffers reacted with the chloramines. To provide constant pH for
the plattnerite experiments with influent solutions that did not have sufficient buffering capacity
provided by DIC or orthophosphate, solutions of either NaOH or HNO3 were added to the influ-
ent from a syringe pump to continuously adjust the pH as necessary to maintain the target pH. For
these experiments the effluent pH was continuously measured using a small flow-through cell with
a pH electrode so that acid or base additions to the influent could be adjusted in near real time.
Each experimental condition was run in duplicate or triplicate with a procedural blank. The
procedural blank consisted of a reactor run under the same conditions as the other reactors, but
without any of the solid added to the system.
Dissolution rates were determined by operating the flow-through reactors for durations
equivalent to 50 or more hydraulic residence times. By continuously flushing the products of
dissolution from the reactor, the effluent dissolved lead concentration approaches a steady-state
concentration that is controlled by the rate of dissolution of the solid phase. Figure 3.9 illustrates
the evolution of the effluent concentration from an initial value (C0) to a steady-state concentra-
tion (Css) that is below the equilibrium solubility of the solid (Ceq). This approach is different from

batch experiments in which the dissolved concentrations increase until they reach equilibrium
solubility; the accumulation of reaction products and the presence of initial labile phases in batch
experiments can complicate the quantification of rates. Ideally in a flow-through experiment the
steady-state effluent concentration will be significantly less than the equilibrium solubility of the
dissolving solid.
For a flow-through reactor the dissolution rate can be quantified by Equation 3.1,
(Ceff − Cinf )
Rate = (3.1)
tres : [solid] : A
where Rate = dissolution rate (mol m–2 min–1)
Ceff = lead concentration in effluent (mol/L)
Cinf = lead concentration in the influent (mol/L)
tres = hydraulic residence time (s)
[solid] = solid concentration in reactor (g/L)
A = specific surface area of the solid, square meters per gram (m2/g).
The steady-state effluent concentration was used in Equation 3.1 for Ceff. This concentration was
determined as the average of the effluent concentration for at least 5 samples that spanned at least
5 residence times. The influent concentration was 0 for all experiments.
A dissolution rate constant k (mol m–2 min–1) can then be calculated from the rate by
accounting for the distance of the actual effluent solution from the predicted equilibrium solubility
of the solid.
Rate = k (1 − 10 SI ) Ω (3.2)
The distance to equilibrium is accounted for by the saturation index (SI) of the solution, which is
defined as the logarithm of the ratio of the ion activity product (IAP) to the solubility product of
the solubility-controlling solid (Ksp) (shown in Equation 3.3 and 3.4 for the cases of hydrocerussite
and hydroxylpyromorphite). The parameter Ω is an empirical coefficient to account for the non-
elementary nature of dissolution reactions, and it was set to 1 in all calculations.
[Pb 2 +] 3 [CO 23 −] 2 [H +] − 2
[Pb 2 +] 5 [PO 34 −] 3 [H +] − 1
Ksp Ksp, hydroxylpyromorphite
Prior to running dissolution experiments a tracer study was conducted to test for well-
mixed conditions in the reactors. Influent with a known concentration of aqueous lead was fed to
the reactors in conditions simulating the dissolution experiment setup. Reactors contained no solid
lead and the initial lead concentration in the reactors was zero for the study. This study confirmed
that conditions within the reactors were well-mixed. See Appendix D for more information.

Table 3.3
Conditions studied for hydrocerussite dissolution
pH
7.5 8.5 10.0
Phosphate Phosphate Phosphate
(mg P/L) (mg P/L) (mg P/L)
0 4 0 4 0 4
Chloramine 0 1 10 2 11 3 12
0 (mg/L as Cl2)
2
Dissolved
Inorganic Chloramine 0 4 13 5 14 6 15
10 (mg/L as Cl2)
Carbon 2 19 24 20 25 21 26
(mg C/L)
Chloramine 0 7 16 8 17 9 18
50 (mg/L as Cl2)
2 22 23
Numbers are experiment identification numbers for each condition. Gray boxes are conditions that were not examined.
RESULTS AND DISCUSSION
Hydrocerussite Dissolution Rates
Overview
Hydrocerussite (Pb3(CO3)2OH(S)) is one of the most frequently observed corrosion prod-

ucts in scales on lead pipes. The dissolution rates of hydrocerussite were quantified at different
influent water conditions. The conditions were chosen to systematically investigate the effects of
water chemistry on dissolution rates (Table 3.3). The dissolution rate was experimentally quanti-
fied through measurement of the effluent lead concentration and hydraulic residence time of the
reactor. The dissolution rate was then normalized to solid surface area because the solid concen-
tration (g/L) and specific surface area (m2/g) are known. A summary of the dissolution rates of
hydrocerussite are found in Table 3.4.
Effect of pH and DIC on Hydrocerussite Dissolution Rates
The effect of dissolved inorganic carbon (DIC) on hydrocerussite dissolution rate was
investigated at pH 7.5, 8.5, and 10.0. The effluent lead concentration from the dissolution reactors
was measured as a function of time for each reactor. To illustrate the overall approach to quantify-
ing the dissolution rate, the results are shown in Figure 3.10 for pH 7.5. For subsequent conditions,
the effluent concentrations versus time will not be shown and only the dissolution rates determined
from the steady-state concentrations will be presented. Steady-state was achieved after 18 resi-
dence times (1080 min). The steady-state concentration of lead (Css) decreased (from 5.60·10–6 to
3.88·10–7 M) as the DIC increased from 0 to 50 mg C/L. These steady-state concentrations were
then used in Equation 3.1 to calculate the dissolution rates (shown in Figure 3.11). The steady-
state concentrations were quite close to the predicted equilibrium concentrations (Ceq) of 8.51·10–7

Table 3.4
Conditions and results of hydrocerussite dissolution experiments
Steady-state
Influent Composition Residence effluent Equilibrium Dissolution Rate
Experiment DIC Orthophosphate Chloramine Measured time lead lead rate × 109 constant × 109
ID* pH (mg/L as C) (mg/L as P) (mg/L as Cl2) pH† (min) (nM)‡ (nM) § (mol/min·m2) (mol/min·m2)**
1A 7.5 0 0 0 7.25 60 5818.30 5836.19 20.46 6672.86
1B 7.5 0 0 0 7.26 60 5547.00 5745.33 19.50 565.00
1C 7.5 0 0 0 7.28 60 5759.90 5569.21 20.25 N/A

2A 8.5 0 0 0 8.59 60 1173.20 1548.07 4.13 17.04
2B 8.5 0 0 0 8.54 60 1354.80 1584.36 4.76 32.88
2C 8.5 0 0 0 8.44 60 1224.20 1668.79 4.30 16.16
3A 10 0 0 0 9.96 60 600.20 2052.20 2.11 2.98
3B 10 0 0 0 9.87 60 519.00 1884.69 1.82 2.52
3C 10 0 0 0 9.79 60 759.20 1762.88 2.67 4.69
4A 7.5 10 0 0 7.36 60 503.10 1274.49 1.77 2.92

4B 7.5 10 0 0 7.40 60 455.80 1227.80 1.60 2.55
4C 7.5 10 0 0 7.42 60 385.30 1205.51 1.35 1.99
5A 8.5 10 0 0 8.58 60 511.10 481.11 1.80 N/A
5B 8.5 10 0 0 8.55 60 610.70 491.40 2.15 N/A
5C 8.5 10 0 0 8.57 60 557.00 484.50 1.96 N/A
6A 10 10 0 0 9.93 60 186.80 254.82 0.66 2.46
6B 10 10 0 0 9.84 60 390.20 259.02 1.37 N/A
6C 10 10 0 0 9.86 60 511.60 258.00 1.80 N/A
7A 7.5 50 0 0 7.39 60 469.50 1990.04 1.65 2.16
7B 7.5 50 0 0 7.43 60 413.00 1928.47 1.45 1.85
7C 7.5 50 0 0 7.57 60 392.70 1731.13 1.38 1.79
8A 8.5 50 0 0 8.50 60 738.80 944.71 2.60 11.92
8B 8.5 50 0 0 8.54 60 862.20 927.12 3.03 43.29
8C 8.5 50 0 0 8.49 60 821.60 949.30 2.89 21.48
(continued)

Table 3.4 (Continued)
Steady-state
ID* pH (mg/L as C) (mg/L as P) (mg/L as Cl2) pH† (min) (nM) ‡ (nM) § (mol/min·m2) (mol/min·m2)**
9A 10 50 0 0 9.85 60 416.70 959.69 1.47 2.59

9B 10 50 0 0 9.83 60 606.00 953.84 2.13 5.84
9C 10 50 0 0 9.78 60 615.90 938.83 2.17 6.30

10A 7.5 0 4 0 7.43 30 5.4078 NC 0.038 N/A
10B 7.5 0 4 0 7.46 30 5.9033 NC 0.041 N/A
11A 8.5 0 4 0 8.28 30 2.5530 NC 0.018 N/A
11B 8.5 0 4 0 8.42 30 1.4977 NC 0.011 N/A
12A 10 0 4 0 9.81 30 2.2483 NC 0.016 N/A
12B 10 0 4 0 9.71 30 4.5105 NC 0.032 N/A
13A 7.5 10 4 0 7.54 30 2.4250 NC 0.017 N/A
13B 7.5 10 4 0 7.52 30 0.4245 NC 0.003 N/A
14A 8.5 10 4 0 8.38 30 0.9916 NC 0.007 N/A
14B 8.5 10 4 0 8.56 30 2.6779 NC 0.019 N/A
15A 10 10 4 0 9.60 30 5.5966 NC 0.039 N/A
15B 10 10 4 0 9.57 30 6.0683 NC 0.043 N/A
16A 7.5 50 4 0 7.75 30 2.8608 NC 0.020 N/A
16B 7.5 50 4 0 7.78 30 2.6479 NC 0.019 N/A
17A 8.5 50 4 0 8.55 30 13.3893 NC 0.094 N/A
17B 8.5 50 4 0 8.60 30 16.4629 NC 0.116 N/A
18A 10 50 4 0 9.65 30 22.6317 NC 0.159 N/A
18B 10 50 4 0 9.72 30 55.4328 NC 0.389 N/A
19A 7.5 10 0 2 7.57 30 589.24 1055.10 2.07 4.69
19B 7.5 10 0 2 7.61 30 564.92 1019.66 1.99 4.45
(continued)

Steady-state
ID* pH (mg/L as C) (mg/L as P) (mg/L as Cl2) pH† (min) (nM)‡ (nM)§ (mol/min·m2) (mol/min·m2)**
20A 8.5 10 0 2 8.50 30 468.98 509.22 1.65 20.87

20B 8.5 10 0 2 8.34 30 508.55 572.21 1.79 16.07
21A 10 10 0 2 10.35 30 540.06 253.63 1.90 N/A
21B 10 10 0 2 10.36 30 523.67 254.25 1.84 N/A
22A 8.5 50 0 2 8.44 30 543.01 944.71 1.91 4.49
22B 8.5 50 0 2 8.45 30 437.22 927.12 1.54 2.91
23A 10 50 0 2 10.14 30 582.89 959.69 2.05 5.22

23B 10 50 0 2 10.04 30 493.18 953.84 1.73 3.59
24A 7.5 10 4 2 7.41 30 11.82 NC 0.083 N/A
24B 7.5 10 4 2 7.48 30 15.42 NC 0.108 N/A
25A 8.5 10 4 2 8.49 30 29.52 NC 0.207 N/A
25B 8.5 10 4 2 8.41 30 23.13 NC 0.162 N/A
26A 10 10 4 2 9.93 30 32.33 NC 0.227 N/A
26B 10 10 4 2 9.88 30 63.41 NC 0.445 N/A
*The ID number corresponds to Table 3.3 and the letters A-C identify replicate experiments.
†The weighted average of the effluent pH during the steady-state period is indicated. NM indicates experiments for which the pH was not measured.
‡BDL = below detection limit of 0.5 μg/L (2.4 nM)
§ NC = equilibrium lead concentrations were not calculated because multiple solids present.
**Rate constants were calculated from the measured dissolution rate and Equation 3.2. N/A indicates that determination of a rate constant was not applicable
because the steady-state effluent concentrations were below detection limits or were not sufficiently below predicted equilibrium concentrations.
14
pH = 7.5, DIC = 0.00 mg/L, Css = 5.60 µM
12 pH = 7.5, DIC = 2.19 mg/L, Css = 1.88 µM

pH = 7.5, DIC = 10.0 mg/L, Css = 0.45 µM
pH = 7.5, DIC = 50.0 mg/L, Css = 0.38 µM
10
Pb Concentration (µM)
ResidenceTime
Figure 3.10 The effect of DIC on effluent lead concentrations from flow-through reactors
loaded with hydrocerussite at pH = 7.5. Steady-state concentrations (Css) are labeled with
dashed lines. Error bars represent 95% confidence intervals for each data point.
M and 1.39·10–6 M for pH 7.5 and 8.5, respectively (details of calculations are in Appendix E).
Equation 3.2 can then be used to calculate the dissolution rate constant, but because of the similar-
ity of equilibrium and steady-state values, the calculation of the saturation index has great uncer-
tainty, and this uncertainty then also applies to the calculated rate constant.
The dissolution rates of hydrocerussite as a function of pH (7.5, 8.5, and 10.0) and DIC (0,
10, and 50 mg C/L) are summarized in Figure 3.11 and Table 3.4. Steady-state lead concentrations
were observed between 12 to 18 residence times. The highest dissolution rate was found at pH
7.5 with no dissolved inorganic carbon concentration in the influent. While the lowest dissolution
rate occurred at a pH of 10 and a DIC of 10 mg C/L. In the case of no DIC in the influent, the dis-
solution rates decreased as a function of pH. The decrease in the dissolution rate is independent
of pH for the influent DIC concentration of 10 and 50 mg C/L. The dissolution rates reach a local
minimum at each pH as DIC increases due to the formation of dissolved lead carbonate complexes
(PbHCO3+, PbCO30, and Pb(CO3)22–) at higher pH values (8.5 and 10).
The steady-state lead concentrations for a few conditions are close to the predicted equi-
librium lead concentrations of pure hydrocerussite (Figure 3.12). In some cases the measured
steady-state lead concentration is greater than the predicted equilibrium concentration or close to
the equilibrium concentration (pH = 7.5 and DIC = 0 mg/L, pH = 8.5 and DIC = 10 mg C/L, and
pH = 10 and DIC = 10 mg C/L). For these three conditions the dissolution rate constant cannot be

Figure 3.11 Dissolution rates of hydrocerussite as a function of pH with influent concentra-

tions of dissolved inorganic carbon (0, 10, and 50 mg C/L)
determined from Equation 3.2 because the measured concentration is not sufficiently far from the
equilibrium concentration. The measured lead concentrations follow the trends of the predicted
equilibrium concentration as a function of pH (Figure 3.12), which suggests that the measured lead
concentrations can be predicted by the equilibrium concentrations. Dissolution is occurring fast
enough that equilibrium concentrations are approached on time-scales faster than the 1 hour resi-
dence time of the reactor. When the dissolution rate constant (mol/min·m2) is plotted as a function
of pH (Figure 3.13) it is apparent that the dissolution rate is independent of pH for DIC conditions
of 10 and 50 mg C/L. As the carbonate concentration for each reaction condition increases, the
reaction rate constant decreases and then plateaus at approximately 10–8.5 (Figure 3.14).
Effect of Phosphate on Hydrocerussite Dissolution
The general strategy of corrosion control is to promote the formation of low solubility lead-
containing solids that passivate the surface of the lead pipe. Phosphate inhibitors can effectively
mitigate lead concentrations in tap water (Edwards et al. 1999; Schock 1999; Churchill et al. 2000;
Cantor et al. 2003). The effect of adding 4 mg P/L orthophosphate to the influent condition on the
overall lead release rate of hydrocerussite was investigated as a function of pH and DIC concentra-
tion. The decrease in dissolved lead concentration was rapid and steady-state lead concentrations
were observed within 10 hydraulic residence times. The overall lead release rate for each condition

-1
DIC = 0 mg/L
-2 DIC = 10 mg/L
DIC = 50 mg/L
DIC = 0 mg/L
log [PbDiss ] (M)
-3
DIC = 10 mg/L
-4
DIC = 50 mg/L
-5
-6
-7
4 5 6 7 8 9 10 11 12
pH
Figure 3.12 Predicted equilibrium dissolved lead concentrations (lines) of 1 g/L hydrocerus-
site as a function of pH and dissolved inorganic carbon concentration. Measured steady state
concentrations are shown with open symbols.
was several orders of magnitude lower than without the addition of orthophosphate (Figure 3.15).
The effects of pH and DIC when orthophosphate is present are shown in Figure 3.16 and can be
compared with the results without orthophosphate in Figure 3.11. When orthophosphate is present,
the presence of DIC and increasing pH causes higher dissolution rates because of the formation of
soluble lead-carbonate complexes.
During the dissolution experiment, the hydrocerussite solid dissolves and hydroxylpyro-
morphite forms. Hydroxylpyromorphite formation was confirmed using powder X-ray diffraction
(Figure 3.17). Imaging with scanning electron microscopy (Figure 3.18) also indicates the forma-
tion of hydroxylpyromorphite following the reaction of hydrocerussite with orthophosphate. The
measured steady-state lead concentrations were much less than the predicted equilibrium solubil-
ity of hydrocerussite. Equilibrium calculations were determined by considering the potential for
precipitation of both hydrocerussite and hydroxylpyromorphite. In the absence of DIC but the
presence of orthophosphate, the effluent compositions were close to the predicted equilibrium sol-
ubility of hydroxylpyromorphite with 4 mg P/L orthophosphate (Figure 3.19). The dissolved lead
concentrations over the time-scale of the flow-through experiments were consequently controlled
by the precipitation of hydroxylpyromorphite from lead released by dissolution of hydrocerussite.
This suggests that most of the hydrocerussite surface is no longer in contact with the water because
of surface growth or precipitation of hydroxylpyromorphite blocking access to the reactive sites.
When both DIC and orthophosphate are present, the dissolved lead concentrations are below the

-5.0
DIC = 0 mg/L
-5.5
DIC = 10 mg/L
logk (mol/min·m ) -6.0 DIC = 50 mg/L
2
-6.5
-7.0
-7.5
-8.0
-8.5
-9.0
-9.5
7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5
pH
Figure 3.13 The reaction rate constant (mol/min·m2) plotted as a function of pH and influent
dissolved inorganic concentration
-5.0
DIC = 0 mg/L
-5.5
DIC = 10 mg/L
-6.0 DIC = 50 mg/L
logk (mol/min·m )
2
-6.5
-7.0
-7.5
-8.0
-8.5
-9.0
-9.5
2-
log [CO3 ]
Figure 3.14 The reaction rate constant (mol/min·m2) plotted as a function of carbonate
(CO32–) concentration and influent dissolved inorganic concentration

-7
0 mg/L orthophosphate
Log Lead Release Rate (mol/min•m2) 4 mg/L as P orthophosphate

-8
-9
-10
-11
-12
DIC=0 DIC=10 DIC=50 DIC=0 DIC=10 DIC=50 DIC=0 DIC=10 DIC=50
pH 7.5 pH 8.5 pH 10.0
Figure 3.15 Dissolution rates of hydrocerussite as a function of pH at a DIC = 0 mg/L with

and without 4 mg P/L orthophosphate. Note that they-axis is on a logarithmic scale.
Figure 3.16 Dissolution rates of hydrocerussite as a function of pH and dissolved inorganic

carbon with an influent orthophosphate concentration of 4 mg P/L

Figure 3.17 Powder X-ray diffraction of hydrocerussite reacted with an influent concentra-
tion of pH = 8.5, DIC = 0 mg/L, and a P = 4 mg/L. Reference patterns for hydrocerussite and
hydroxylpyromorphite are included.
A B

Figure 3.18 Scanning electron microscopy image of (A) hydrocerussite reacted at pH = 10
and DIC = 10 mg/L and (B) hydrocerussite reacted with a pH = 8.5 and P = 4 mg/L after
10 minutes in a batch reactor

-3
DIC = 0 mg/L
-4 DIC = 10 mg/L
DIC = 50 mg/L
-5
log [Pbdiss ] (M)
-6 DIC = 50 mg/L
DIC = 10 mg/L
-7
DIC = 0 mg/L
-8
-9
-10
4 5 6 7 8 9 10 11 12
pH
Figure 3.19 Predicted dissolved lead concentrations (lines) in equilibrium with hydroxyl-
pyromorphite with an influent orthophosphate concentration of 4 mg P/L. Measured steady-
state effluent concentrations are shown with open symbols.
equilibrium solubility of hydroxylpyromorphite. For these conditions, the orthophosphate may be

inhibiting hydrocerussite dissolution by adsorbing to the surface and blocking sites of dissolution.
Effect of Chloramines on Hydrocerussite Dissolution
The effect of 2 mg/L as Cl2 monochloramine was investigated for pH values of 7.5, 8.5, and
10 and DIC concentrations of 10 and 50 mg C/L (Figure 3.20). The dual effect of monochloramine
and phosphate was investigated at pH values of 7.5, 8.5, and 10 and a DIC concentration of 10 mg
C/L. Monochloramine stock solutions (200 mg/L as Cl2) were prepared immediately before each
dissolution experiment. A 0.79 Cl2:N molar ratio was used to minimize the formation of dichlora-
mine. No buffering solutions were used due to the reaction of monochloramines with the organic
buffer. Although chloramines are not expected to affect the dissolution rate of hydrocerussite, the
ammonia needed to produce monochloramine may cause a slight increase in lead dissolution due
to the formation of a soluble lead ammonia complex (Equation 3.5) (Martell and Smith 1974).
Pb2+ + NH3 = PbNH32+ Log K = 1.90 (3.5)
This is a relatively weak complex (e.g., as compared to those with the carbonate ion) and it can
only have a slight effect. The dissolution rates in the presence of 2 mg/L as Cl2 monochloramine
were about the same as the dissolution rates in the absence of monochloramine (Figure 3.20).
However, the lead release rates with both monochloramine and orthophosphate concentrations

5.0
0 mg/L chloramines
4.5 2 mg/L chloramines
Lead Release Rate x 10 (mol/min•m )

2 4.0
3.5
3.0
9
2.5
2.0
1.5
1.0
0.5
0.0
pH 7.5 pH 8.5 pH 8.5 pH 10 pH 10
DIC = 10 DIC = 10 DIC = 50 DIC = 10 DIC = 50
Figure 3.20 Dissolution rates of hydrocerussite as a function of pH with and without 2 mg/L
as Cl2 monochloramine
were higher than the rates with just orthophosphate (Figure 3.21). This may be caused by the inter-
ference of the lead(II) ammonia complex with the formation of hydroxylpyromorphite.
Summary of Hydrocerussite Dissolution
Hydrocerussite dissolution rates are affected by pH and the concentrations of DIC and
orthophosphate. Dissolution rates decrease with increasing pH for systems containing no ortho-
phosphate. With increasing DIC concentration, the dissolution rate first decreases and then slightly
increases. The effect of DIC is caused by the concentration of the carbonate ion, which decreases
the dissolution rate of hydrocerussite by lowering the equilibrium lead concentration and increases
the dissolution rate through the formation of soluble lead(II)-carbonate complexes. Ultimately the
effect of lowering the solubility of hydrocerussite is much more significant.
Steady-state effluent lead concentrations are very close to the predicted equilibrium solu-
bility of hydrocerussite. This suggests that dissolved lead concentrations are controlled by the
equilibrium and not the kinetics of hydrocerussite dissolution. Consequently, equilibrium calcula-
tions can provide good estimates of dissolved lead concentrations in lead service lines with hydro-
cerussite as the dominant component of the corrosion products. Kinetics may play a role during
conditions of flowing water, but for stagnation times of one hour or more, equilibrium estimations
can provide good estimates of dissolved lead.
The addition of orthophosphate dramatically lowers the net release rate of lead during
hydrocerussite dissolution. Dissolution rates are more than two orders of magnitude lower in the

0.40
0 mg/L chloramines
2 mg/L chloramines
0.35
Lead Release Rate x 109 (mol/min•m2)

0.30
0.25
0.20
0.15
0.10
0.05
0.00
pH 7.5, P = 4 pH 8.5, P = 4 pH 10, P = 4
Figure 3.21 Dissolution rates of hydrocerussite as a function of pH with and without 2 mg/L
as Cl2 monochloramine in the presence of 4 mg P/L orthophosphate and 10 mg C/L DIC
presence of orthophosphate than in the absence. The decrease is caused by the formation of the
lead phosphate solid hydroxylpyromorphite. The majority of the lead released from hydrocerussite
is sequestered by precipitation in hydroxylpyromorphite. This lead phosphate phase was directly
observed by SEM and XRD, and several conditions have dissolved lead concentrations that are
consistent with the equilibrium solubility of hydroxylpyromorphite. As expected, the dissolution
rates increase with increasing DIC because of the formation of soluble lead(II)-carbonate com-
plexes, but even with these complexes the steady-state effluent lead concentrations are far below
those in the absence of orthophosphate.
Plattnerite Dissolution Rates
Plattnerite is an important lead corrosion product in systems that have maintained a high
concentration of free chlorine as a residual disinfectant. The dissolution rate of this solid is par-
ticularly relevant to assessing the rates of lead release when the concentration or type of the
residual disinfectant is changed. When plattnerite is stable at the high oxidation-reduction poten-
tial provided by free chlorine, dissolved lead concentrations were maintained at low levels. When
the free chlorine is not present, the dissolution rate of plattnerite is anticipated to be controlled by
the water chemistry. A series of 21 experiments (Table 3.5) were performed to evaluate the impact
of pH, dissolved inorganic carbon, orthophosphate, and chloramines on the dissolution rate of
plattnerite. DIC and pH were expected to have the greatest effect on the dissolution rate and hence

Table 3.5
Conditions studied for plattnerite dissolution
pH
7.5 8.5 10.0
Orthophosphate Orthophosphate Orthophosphate
(mg P/L) (mg P/L) (mg P/L)
0 1 0 1 0 1
Chloramine 0 1 10 2 11 3
0 (mg/L as Cl2)
2 14 18 15 19
Dissolved
Inorganic Chloramine 0 4 5 6
10 (mg/L as Cl2)
Carbon 2
(mg/L as C)
50 (mg/L as Cl2)
2 16 20 17 21
were examined at three values. The effects of chloramines and orthophosphate were explored
using simple absence-presence tests.
The conditions and results of all plattnerite dissolution experiments, including replicate
experiments, are summarized in Table 3.6. The experiment ID numbers correspond to the condi-
tions given in Table 3.5. In general the duplicate experiments conducted at each condition were in
agreement; the largest factors of differences were for the lowest overall lead concentrations, which
were near the detection limit of lead by ICP-MS. Variability was also introduced by the chal-
lenges of maintaining a constant pH in the absence of a pH buffer; pH buffers were not used in the
plattnerite experiments because of their ability to accelerate the dissolution of plattnerite by acting
as reductants. The weighted averages of the pH over the period for the steady-state lead effluent
concentration are noted in the table.
The dissolution rate of plattnerite is examined as a function of water chemistry in the
following sections. Steady-state effluent concentrations are compared with predicted equilibrium
concentrations to assess whether dissolved lead is controlled by kinetic or equilibrium processes.
Characterization of the solids remaining after the dissolution experiments did not identify the
precipitation of any secondary phases. Finally an empirical model is presented that examines the
variation in dissolution rate with all four of the water chemistry parameters evaluated.
Effect of pH and DIC on Plattnerite Dissolution
Dissolution rates. The dissolution rate decreased as the pH increased from 7.5 to 8.5 when
other parameters were held constant for almost all experiments (Figure 3.22). An exception was
for the condition of 50 mg C/L DIC with orthophosphate and choramines present; for this condi-
tion the dissolution rate was not significantly affected by pH. For experiments without orthophos-
phate and chloramines, the effect of further increasing the pH 10 was also examined. The increase
from pH 8.5 to 10.0 resulted in very slight increases in the dissolution rate at 0 and 10 mg C/L DIC
and to a decrease for 50 mg C/L DIC.

Table 3.6
Conditions and results of plattnerite dissolution experiments
Steady-state
Experiment DIC Orthophosphate Chloramine Measured time lead lead rate × 10 9 constant × 10 9
ID* pH (mg/L as C) (mg/L as P) (mg/L as Cl2) pH† (min) (nM) (nM)‡ (mol/min·m2) (mol/min·m2)§
1A 7.5 0.0 0.0 0.0 7.58 30 2 2630 0.02 0.02

1B 7.5 0.0 0.0 0.0 7.52 30 12 2630 0.11 0.11
2A 8.5 0.0 0.0 0.0 8.57 30 6 123 0.05 0.06
2B 8.5 0.0 0.0 0.0 8.58 30 1 123 0.01 0.01
3A 10.0 0.0 0.0 0.0 9.94 30 6 13 0.06 0.11
3B 10.0 0.0 0.0 0.0 9.98 30 3 13 0.03 0.04
3C 10.0 0.0 0.0 0.0 9.78 30 6 13 0.05 0.10
4A 7.5 10.0 0.0 0.0 7.78 30 28 6761 0.26 0.26
4B 7.5 10.0 0.0 0.0 7.78 30 23 6761 0.21 0.21
5A 8.5 10.0 0.0 0.0 8.27 30 15 501 0.14 0.14
5B 8.5 10.0 0.0 0.0 8.26 30 12 501 0.12 0.12
6A 10.0 10.0 0.0 0.0 9.95 30 15 27 0.14 0.33
6B 10.0 10.0 0.0 0.0 9.97 30 22 27 0.20 1.07
7A 7.5 50.0 0.0 0.0 7.62 30 84 23442 0.79 0.79
7B 7.5 50.0 0.0 0.0 7.68 30 83 23442 0.78 0.78
8A 8.5 50.0 0.0 0.0 8.36 30 42 2344 0.39 0.40
8B 8.5 50.0 0.0 0.0 8.42 30 60 2344 0.56 0.58
9A 10.0 50.0 0.0 0.0 9.94 30 27 214 0.25 0.29
9B 10.0 50.0 0.0 0.0 9.99 30 28 214 0.26 0.30
(continued)

Steady-state
9
Experiment DIC Orthophosphate Chloramine Measured time lead lead rate × 10 constant × 10 9
ID* pH (mg/L as C) (mg/L as P) (mg/L as Cl2) pH† (min) (nM) (nM)‡ (mol/min·m2) (mol/min·m2)§
10A 7.5 0.0 1.0 0.0 7.20 30 17 2630 0.16 0.16
10B 7.5 0.0 1.0 0.0 7.11 30 31 2630 0.29 0.30

11A 8.5 0.0 1.0 0.0 8.68 30 8 123 0.08 0.08
11B 8.5 0.0 1.0 0.0 8.34 30 8 123 0.08 0.08
12A 7.5 50.0 1.0 0.0 7.72 30 71 23442 0.67 0.67
12B 7.5 50.0 1.0 0.0 7.70 30 62 23442 0.58 0.58
13A 8.5 50.0 1.0 0.0 8.50 30 25 2344 0.23 0.23
13B 8.5 50.0 1.0 0.0 8.55 30 13 2344 0.12 0.12

14A 7.5 0.0 0.0 2.0 7.30 30 37 0.35 N/A
14B 7.5 0.0 0.0 2.0 7.58 30 8 0.08 N/A
15A 8.5 0.0 0.0 2.0 8.85 30 1 0.01 N/A
15B 8.5 0.0 0.0 2.0 8.69 30 2 0.01 N/A
16A 7.5 50.0 0.0 2.0 7.56 30 77 0.73 N/A
16B 7.5 50.0 0.0 2.0 7.67 30 57 0.54 N/A
17A 8.5 50.0 0.0 2.0 8.53 30 14 0.13 N/A
17B 8.5 50.0 0.0 2.0 8.52 30 18 0.17 N/A
18A 7.5 0.0 1.0 2.0 7.42 30 15 0.15 N/A
18B 7.5 0.0 1.0 2.0 7.45 30 15 0.14 N/A
18C 7.5 0.0 1.0 2.0 7.64 30 15 0.15 N/A
(continued)

Steady-state
Experiment DIC Orthophosphate Chloramine Measured time lead lead rate × 10 9 constant × 10 9
† (mol/min·m2) (mol/min·m2)§
ID* pH (mg/L as C) (mg/L as P) (mg/L as Cl2) pH (min) (nM) (nM)‡
19A 8.5 0.0 1.0 2.0 8.44 30 8 0.08 N/A
19B 8.5 0.0 1.0 2.0 8.07 30 8 0.07 N/A
20A 7.5 50.0 1.0 2.0 7.66 30 16 0.15 N/A
20B 7.5 50.0 1.0 2.0 7.65 30 15 0.15 N/A
21A 8.5 50.0 1.0 2.0 8.52 30 16 0.15 N/A
21B 8.5 50.0 1.0 2.0 8.53 30 18 0.17 N/A
‡NC = Not Calculated. Equilibrium lead concentrations were not calculated for influents containing chloramines because of the uncertainty of the
equilibrium constants for the associated reactions.
§Rate constants were calculated from the measured dissoluton rate and Equation 3.2. N/A indicates that determation of a rate constant was not applicable
because the a predicted equilibrium dissolved lead concentration could not be calculated.

9
pH 7.5
Dissolution Rates x 10 10(mol/m2-min)
8 pH 8.5
pH 10.0
7
0
DIC=0, DIC=10, DIC=50, DIC=0, DIC=0, DIC=0, DIC=50, DIC=50, DIC=50,
P=0, P=0, P=0, P=1, P=0, P=1, P=0, P=1, P=1,
CA=0 CA=0 CA=0 CA=0 CA=2 CA=2 CA=2 CA=0 CA=2
Figure 3.22 Plattnerite dissolution rate as a function of pH and DIC. Conditions are listed
on the x-axis with P denoting the orthophosphate concentration in mg P/L and CA denoting
the chloramines concentration in mg/L as Cl2.
Dissolution rates increased with increasing DIC for all conditions studied (Figure 3.23).
The highest plattnerite dissolution rates in the entire study were observed with 50 mg C/L DIC.
The formation of soluble complexes of lead(II) with carbonate and bicarbonate ions explains the
accelerated dissolution in the presence of DIC. The dissolved lead concentration is a function of
both the dissolution rate and the reaction time, and for reaction times longer than the 30 minute
residence time in the flow-through experiments, dissolved lead concentrations will be higher.
Characterization of the solids remaining at the conclusion of the experiments with influents
containing only DIC found no evidence of transformation of solid into secondary phases. The only
solid observed using X-ray diffraction was plattnerite.
For comparable conditions, the dissolution rates of plattnerite are about two orders of mag-
nitude less than those of hydrocerussite. Plattnerite is much more effective at maintaining low
steady-state dissolved lead concentrations. The influence of dissolved lead-carbonate complexes
on increasing the dissolution rate was apparent for both solids.
Equilibrium versus kinetic control of dissolved lead concentrations. For the influents
with only DIC in the influent, the steady-state concentrations were well below the equilibrium
solubility of any possible solids at all three pH values investigated (Figure 3.24). Solids considered
were the initial plattnerite solid as well as the lead(II) carbonate solids cerussite and hydrocerussite
that will ultimately govern the dissolved lead concentration as the lead(IV) in plattnerite is reduced
and released to solution. The undersaturation of the solution with respect to all possible solids is
consistent with the X-ray diffraction results that observed only plattnerite in solid samples col-
lected at the conclusion of dissolution experiments.

9
Dissolution Rates x 10 (mol/m -min)
DIC = 0
8
DIC = 10 mg/L
2
7 DIC = 50 mg/L
6
10
0
pH=7.5, pH=7.5, pH=7.5, pH=7.5, pH=8.5, pH=8.5, pH=8.5, pH=8.5, pH=10,
CA=0, CA=0, CA=2, CA=2, CA=0, CA=0, CA=2, CA=2, CA=0,
P=0 P=1 P=0 P=1 P=0 P=1 P=0 P=1 P=0
Figure 3.23 Effect of DIC on plattnerite dissolution rates. Conditions are listed on the x-axis
with P denoting the orthophosphate concentration in mg P/L and CA denoting the chlora-
mines concentration in mg/L as Cl2.
Dissolution in the flow-through reactors was occurring at conditions that are far from equi-
librium. For such conditions, the Gibbs free energy of reaction (ΔG) is very negative. The Gibbs
free energy can be related to the saturation index (Equation 3.6), and Equation 3.2 is one form of
the effect of the Gibbs free energy on the reaction rate (Equations 3.7 and 3.8).
ΔG = RT ln d IAP n = 2.303RT log d IAP n = 2.303RT $ SI (3.6)

Ksp Ksp
where R is the ideal gas constant (8.314 J mol–1 K–1) and T is absolute temperature (K).
ΔG
f (ΔG) = (1 − e RT ) Ω = (1 − 10 SI ) Ω (3.7)
Rate = k $ f (ΔG) (3.8)
For the conditions with plattnerite, f(ΔG) is close to 1 and Equation 3.2 can be simplified to
Equation 3.9.
Rate = k $ 1 = k (3.9)
In this manner, the dissolution rates and the rate constants are the same. For the steady state opera-
tion of the reactors, the dissolution reaction is kinetically limited.

-1 DIC = 0 mg C/L
DIC = 10 mg C/L
-2
DIC = 50 mg C/L
-3 DIC = 50 mg/L
log [Pb]diss (M)
-4 DIC = 10 mg/L
-5
-6 DIC = 0 mg/L
-7
-8
-9
-10
6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0
pH
Figure 3.24 Steady-state and predicted equilibrium concentrations for plattnerite dissolution
as a function of pH and DIC. Solid lines indicate the predicted dissolved lead concentration
in equilibrium with plattnerite. Points identify the steady-state effluent lead concentrations
at specific water chemistry conditions.
Although the increase in DIC leads to an increase in the predicted equilibrium dissolved
lead concentration, this change in the equilibrium concentration is not the cause of the faster disso-
lution with DIC. Rather, the reaction was sufficiently far from equilibrium at pH 7.5 and 8.5 even
in the absence of DIC so that f(ΔG) is approximately 1 both with and without DIC. Consequently,
the presence of DIC leads to a greater value for the rate constant (k) and indicates that the effect
of DIC is to accelerate the rate of the reaction occurring on the surface of the solid. The experi-
ment at pH 10 with 10 mg C/L DIC is the only one for which steady-state concentrations were
close to equilibrium predictions and for which solubility may be controlled by equilibrium. The
predicted equilibrium total lead concentration from only plattnerite increases with increasing DIC
concentration.
Dissolution of plattnerite at conditions without free chlorine to maintain the stability of
lead(IV) occurs through steps of reduction and dissolution, but the sequence of these two steps is
unknown. A conceptual model showing both possible sequences is presented in Figure 3.25. The
release of Pb(IV) to solution followed by the reduction to dissolved Pb(II) is shown as mechanism
C, and mechanism A illustrates the reduction of Pb(IV) to Pb(II) at the surface of the solid followed
by the release of the Pb(II) to solution. The acceleration of the dissolution rate by the presence of
carbonate provides evidence for the pathway involving reduction at the plattnerite surface. The
formation of soluble lead(II)-carbonate complexes can accelerate dissolution by extracting lead(II)
that had formed on the solid surface; there are not soluble lead(IV)-carbonate complexes, and were

A. Reduction at surface with dissolution and precipitation

influenced by carbonate and orthophosphate
dissolution accelerated Pb(II)diss
by CO32-
reductant
PbO2(s)-Pb(II)surf PO43-
reduction
precipitation
Pb(II) Phosphates
PbO2(s)
B. Orthophosphate adsorption blocks surface sites
Pb(IV) oxides
reductant adsorption blocks
PbO2(s)-Pb(II)surf dissolution sites PO 3-
4
reduction
C. Dissolution as Pb(IV) and reduction to Pb(II) in the

aqueous phase
reductant
Pb(IV)diss Pb(II)diss
dissolution reduction
Figure 3.25 Conceptual model of PbO2 dissolution showing three different possible mecha-
nisms and the potential roles of carbonate and orthophosphate
reduction to be occurring in solution, then the addition of DIC would not have changed the overall
dissolved lead concentration. The dissolution of plattnerite is kinetically limited at all, or nearly
all, of the combination of pH and DIC studied and the rate-limiting step appears to be the detach-
ment of lead(II) from the plattnerite surface.
Effect of Phosphate on Plattnerite Dissolution
Dissolution rates. At almost all conditions studied, the addition of 1 mg P/L orthophos-
phate inhibited plattnerite dissolution (Figure 3.26). For the few conditions for which dissolution
rates in Figure 3.26 are seen to be higher with orthophosphate, the increase can be explained
by differences in pH between the conditions with and without orthophosphate; consequently, the
effect is actually caused by pH and not by the orthophosphate (complete pH and dissolution rate
information is included in Table 3.6). For example, at the condition of pH 7.5 with no chloramines
or DIC, the average pH values at steady state were 7.58 and 7.52 for the two replicates without
orthophosphate and 7.20 and 7.12 for the two replicates with orthophosphate. The conditions of
pH 8.5 without DIC and with and without chloramines had similar situations; the buffering pro-
vided by DIC made the pH less variable than in the absence of DIC.
With the increase in dissolution rate with increasing DIC and the decrease with addition
of orthophosphate, the degree to which these effects are offsetting can be assessed by examining
those conditions in Figure 3.26 for which both DIC and orthophosphate increased while the pH
and chloramines concentration remained the same. For the conditions without chloramines at both
pH 7.5 and 8.5, the effect of simultaneously increasing orthophosphate from 0 to 1 mg P/L and
DIC from 0 to 50 mg C/L was a net increase in the dissolution rate, which indicates that the effect
of DIC outweighed that of the orthophosphate. When chloramines were present in the influent, the
net effect of increasing DIC and orthophosphate on the plattnerite dissolution rate was a decrease

9
Dissolution Rate x 10 (mol/m -min) P = 0 mg/L
8 P = 1 mg/L
2
6
10
0
pH=7.5, pH=7.5, pH=8.5, pH=8.5, pH=7.5, pH=7.5, pH=8.5, pH=8.5,
CA=0, CA=0, CA=0, CA=0, CA=2, CA=2, CA=2, CA=2,
DIC=0 DIC=50 DIC=0 DIC=50 DIC=0 DIC=50 DIC=0 DIC=50
Figure 3.26 Effect of orthophosphate on dissolution rates of plattnerite. Conditions are listed
on the x-axis with CA denoting the chloramines concentration in mg/L as Cl2 and DIC denot-
ing the dissolved inorganic carbon concentration in mg C/L.
at pH 7.5 and an increase at pH 8.5. The greater effect of DIC at pH 8.5 than 7.5 is likely the result
of a higher concentration of CO32–, which is the most active species in forming the soluble lead(II)-
carbonate complexes that result in greater dissolution.
The combined effects of altering pH and orthophosphate can also be assessed by selec-
tively examining results presented in Figure 3.26. Decreasing pH had been shown to increase
dissolution rates and increasing orthophosphate to decrease rates. When both of these changes are
made together, the net effect is generally an increase in the dissolution rate, which suggests the
strong importance of pH on controlling dissolution. Only at the condition of 50 mg C/L DIC and
2 mg/L as Cl2 of chloramines was the net effect of increasing phosphate concentration and decreas-
ing pH not significant.
The inhibition of plattnerite dissolution by orthophosphate may be caused by the adsorp-
tion of orthophosphate to the plattnerite surface to block sites of reduction or dissolution or by
the sequestration of lead released from plattnerite by the formation of a secondary precipitate
such as the lead(II) phosphate hydroxylpyromorphite (mechanisms A and B in Figure 3.25). The
characterization of the solids following reaction in the presence of orthophosphate did not pro-
vide any evidence of hydroxylpyromorphite precipitation or of the formation of any other solids
(Figure 3.27). However, the lead that would have been released from the plattnerite to be available
for hydroxylpyromorphite precipitation would be limited by the plattnerite dissolution rate, and
hydroxylpyromorphite may be present at amounts below the detection limit of XRD, which is in
the range of 2–5% by mass.
Equilibrium versus kinetic control of dissolved lead concentrations. The steady-state
effluent lead concentrations were around two orders of magnitude lower than the predicted lead

pH 8.5 DIC 50 P 1
Relative Intensity
pH 8.5 DIC 50 P 1
pH 8.5 DIC 0 P 1
pH 8.5 DIC 0 P 1
Plattnerite PDF#00-025-0447
Scrutinyite PDF#04-011-0549
20 25 30 35 40 45 50 55
2θ (°)
Figure 3.27 X-ray diffraction patterns of solids remaining following flow-through dissolu-
tion experiments of plattnerite under conditions of 1 mg P/L orthophosphate, pH 8.5, and
DIC of 0 or 50 mg C/L
concentrations in equilibrium with plattnerite (Figure 3.28). The thermodynamics of the dissolu-

tion reaction drive plattnerite dissolution. The equilibrium solubility of hydroxylpyromorphite,
which could have formed in the presence of orthophosphate, is lower than that of plattnerite at
both pH 7.5 and 8.5. The steady-state concentrations during plattnerite dissolution were similar
to the concentrations that would be in equilibrium with hydroxylpyromorphite, which indicates
that hydroxylpyromorphite probably formed and that its precipitation took up lead released dur-
ing plattnerite dissolution to result in the lower steady-state effluent lead concentrations. Although
X-ray diffraction analysis could not identify hydroxylpyromorphite formation, the dissolution
rates of plattnerite were probably too small to generate sufficient dissolved lead for the precipita-
tion of hydroxylpyromorphite amounts that would be detectable by XRD. This could explain the
impact of orthophosphate on the dissolution of plattnerite. However, at conditions with 50 mg C/L
DIC, the steady-state concentrations did not reach the equilibrium concentrations of hydroxylpyro-
morphite but orthphosphate still reduced the dissolution rates. Several reasons for this observation
are possible. First, adsorbed phosphate at the plattnerite surface could occupy the dissolution sites,
resulting in a lower dissolution rate. Second, the local concentrations of orthophosphate and lead
at the plattnerite surface could be higher than the bulk concentration due to factors like adsorption
and make the local environment oversaturated with respect to hydroxylpyromorphite.

0
DIC = 0 mg C/L
-1
DIC = 50 mg C/L
-2
-3
log [Pb]diss (M)
-4
-5
-6 DIC = 50 mg/L
-7
-8 DIC = 0 mg/L
-9
6.0 7.0 8.0 9.0 10.0 11.0
pH
Figure 3.28 Steady-state and predicted equilibrium concentrations for plattnerite dissolution
in the presence of 1 mg P/L orthophosphate. Solid lines indicate the predicted equilibrium
concentration as controlled by the solubility of plattnerite and/or hydroxylpyromorphite.
Points identify the steady-state effluent lead concentrations at specific water chemistry con-
ditions.
Effect of Monochloramine on Plattnerite Dissolution
Chloramines have an inhibitory effect on the dissolution of plattnerite at almost all condi-
tions for which other parameters were held constant (Figure 3.29). For most conditions, 2 mg/L
as Cl2 chloramines reduced the dissolution rate significantly, compared with that of the same con-
dition but without chloramines. The condition of pH 7.5, no orthphosphate and no DIC was the
exception, which again could be caused by the difference of average steady-state pH between the
conditions without and with chloramines (7.52 and 7.58 without chloramines and 7.30 with chlora-
mines). The enhancing effect of pH could mask the inhibiting effect of chloramines. Chloramines
inhibit dissolution more strongly in the presence of 50 mg C/L DIC than in the absence of DIC.
Equilibrium calculations using the reactions and equilibrium constants included in
Appendix B are consistent with less dissolution in the presence of chloramines as a result of a
higher oxidation-reduction potential. However, due to uncertainty in the equilibrium constants for
chloramines species, exact equilibrium predictions in the presence of chloramiens could not be
made. Chloramines may be able to delay the reduction of the lead(IV) in plattnerite, but previous
research found that chloramines were not capable of oxidizing lead(II) to lead(IV) (Rajasekharan
et al. 2007). Further, recent work by Lin and Valentine (2008b) found that chloramines could actu-
ally increase the reduction of lead(IV) oxide solids, which is not consistent with the observations

9
Dissolution Rate x 10 10(mol/m2-min)
CA = 0 mg/L
8
CA = 2 mg/L
7
0
pH=7.5, pH=7.5, pH=8.5, pH=8.5, pH=7.5, pH=7.5, pH=8.5, pH=8.5,
P=0, P=0, P=0, P=0, P=1, P=1, P=1, P=1,
DIC=0 DIC=50 DIC=0 DIC=50 DIC=0 DIC=50 DIC=0 DIC=50
Figure 3.29 Effect of chloramines on dissolution rates of plattnerite. Conditions are listed on
the x-axis with P denoting the orthophosphate concentration in mg P/L and DIC denoting the
dissolved inorganic carbon concentration in mg C/L.
from the flow-through experiments in the present project. The differences between the results
of Lin and Valentine and the present study may be explained by differences in the nature of the
experiments (flow-through versus batch), the reaction times (much shorter in the present work),
and differences in exact water chemistry conditions. In fact, for selected experiments when the
flow-through reactors were operated in a batch mode (no flow) for 24 hours following the period
with flow, the concentration of chloramines decreased dramatically and the dissolved lead con-
centration increased, which is consistent with the work of Lin and Valentine that chloramines can
promote plattnerite dissolution during batch dissolution.
The combined effects of pH, DIC, and chloramines were examined by considering the
simultaneous changes in parameters that would independently have opposite effects. A decrease
in pH from 8.5 to 7.5 would increase the rate, and an increase in chloramines would decrease the
rate. When these changes were made together the net result was usually an increase in the rate,
which indicates again a dominant role of pH in controlling plattnerite dissolution rates. When DIC
and chloramines are both increased relative to a base condition, the net result is an increase in the
dissolution rate, which suggests that DIC plays a more important role than chloramine in control-
ling the dissolution rate.
Empirical Model for Plattnerite Dissolution Rates
Two empirical models were evaluated with respect to their ability to simulate the experi-
mental dissolution rates as a function of the four water chemistry parameters evaluated. These

8
1:1 line
Model Rate x 1010(mol/m2-min)

7
-1
10 2
Experimental Rate x 10 (mol/m -min)
Figure 3.30 Empirical linear model of plattnerite dissolution rates. The 1:1 line is shown for
reference.
models are empirical because they are based solely on multivariate regression and do not consider
specific dissolution mechanisms. These models were attempted to assess the ability of simple
equations to reflect experimentally-measured rates.
A linear model was formulated using Equation 3.10
R = a×pH + b×DIC + c×[P] + d×[CA] +e (3.10)
Where, R is the dissolution rate and a, b, c, d, and e are coefficients associated with the four water
chemistry parameters studied, and [P] and [CA] represent the concentrations of orthophosphate
and chloramines, respectively. A least squares optimization of Equation 3.9 to the experimen-
tal data yielded coefficients of –0.12·10–9, 0.0055·10–9, –0.11·10–9, –0.53·10–9, and 1.2·10–9 mol
min–1 m–2 for coefficients a–e. The measured and modeled rates are compared in Figure 3.30.
An empirical exponential model (Equation 3.11 and Equation 3.12 in logarithmic form)
was tried as an alternative to the linear model,
R = e’×[H+]a’×[DIC]b’×[PO43–]c’×[CA]d’ (3.11)
logR = a’×(–pH) + b’×log[DIC] + c’×log[P] + d’×log[CA] + loge’ (3.12)

80
(mol/m -min)
1:1 line
70
2
60
50
11
Model Constant x 10
40
30
20
10
11 2
Experimental Rate Constant x 10 (mol/m -min)
Figure 3.31 Empirical exponential model of plattnerite dissolution rate constants. The 1:1
line is shown for reference.
where a’–e’ are again coefficients. Least squares optimization of the data yield coefficients of 0.28,
0.11, 0.021, 0.013, and 10–7.52 for coefficients a’–e’, respectively. The overall fit is better for the
linear model than for the exponential model, which is shown in Figure 3.31.
Summary of Plattnerite Dissolution Rates
The dissolution rate of plattnerite was strongly controlled by important water chemis-
try parameters. The rate of dissolution increased with decreasing pH and with increasing DIC.
Dissolution rates were more than two orders of magnitude lower than those of hydrocerussite at
comparable conditions. The effect of DIC is consistent with a dissolution mechanism explained by
the formation of soluble lead(II)-carbonate complexes that accelerate the release of lead(II) formed
by reduction of lead(IV) on the plattnerite surface. In the absence of orthophosphate, dissolved
lead concentrations were controlled by the dissolution rate of plattnerite and not by its equilibrium
solubility. While most steady-state effluent concentrations for a 30 minute residence time were
low, concentrations would increase for longer contact times of plattnerite with the water. Of the
four water chemistry parameters investigated, pH had the greatest effect on dissolution rates.
Addition of orthophosphate significantly decreased the net rate of lead release during
plattnerite dissolution. The decrease is likely caused by a combination of mechanisms of precipita-
tion of the lead(II) phosphate solid hydroxylpyromorphite and by adsorption of phosphate to the
plattnerite surface to block sites of dissolution. In the absence of DIC, steady-state lead concentra-
tions were consistent with control by equilibrium with hydroxylpyromorphite, although insufficient

Table 3.7
Conditions studied for hydroxylpyromorphite dissolution
pH
7.5 8.5 10.0
DIC (mg C) DIC (mg C/L) DIC (mg C/L)
0 50 0 50 0 50
0
(mg/L as Cl2) 2 12 14 18 20
Orthophosphate Chloramine 0 2 11 5 17 8
1
(mg P/L) (mg/L as Cl2) 2 13 15 19 21
Chloramine 0 3 6 9
4
(mg/L as Cl2) 2
amounts of this phase may have formed to be detectable by X-ray diffraction. Chloramines also
may have an inhibitory effect on plattnerite dissolution, but the magnitude of this effect is lower
than that of orthophosphate addition.
Two empirical models were proposed to fit the experimental data. These models provide
reasonable fits of the results and may be useful for prediction of dissolution rates from plattnerite
at various water chemistry conditions.
Hydroxylpyromorphite Dissolution Rates
Hydroxylpyromorphite can be an important lead corrosion product for systems using ortho-
phosphate to mitigate lead release to the water. As previously noted, hydroxylpyromorphite or a
closely related phase formed during the dissolution of the lead(II) carbonate hydrocerussite in the
presence of orthophosphate. The decrease in lead release rates during plattnerite dissolution that
accompanied the addition of orthophosphate may also be related to the formation of hydroxylpyro-
morphite. To assess the influence of water chemistry on the stability of hydroxylpyromorphite, a
set of 21 experiments was conducted that examined various combinations of pH and concentra-
tions of DIC, orthophosphate, and chloramines (Table 3.7).
The conditions and results of all hydroxylpyromorphite dissolution experiments, including
replicate experiments, are summarized in Table 3.8. The experiment ID numbers correspond to the
conditions given in Table 3.7. The inclusion of replicate experiments (either duplicates or tripli-
cates) can be used to assess the variability in dissolution rate measurements for a given condition.
Steady-state effluent lead concentrations for replicates varied by as much as a factor of two, but
differences between conditions were still clear for most conditions. The variability among repli-
cates was greater for hydroxylpyromorphite than for the hydrocerussite and the plattnerite, which
may have been due to the overall lower concentrations of lead in the experiments.
The following sections examine the influence of each individual parameter (pH, orthophos-
phate, DIC, and chloramines) on the dissolution rate as well as the relative impacts of each param-
eter. The solids were characterized after selected experiments. Finally, the net dissolution rates are

Table 3.8
Conditions and results of hydroxylpyromorphite dissolution experiments
Steady-state
ID* pH (mg/L as C) (mg/L as P) (mg/L as Cl2) pH† (min) (nM)‡ (nM) (mol/min·m2) (mol/min·m2)§
1A 7.5 0 0 0 7.59 30 91 220 0.190 0.190
1B 7.5 0 0 0 7.61 30 97 220 0.202 0.202
1C 7.5 0 0 0 7.39 30 94 220 0.196 0.196

2A 7.5 0 1 0 7.54 31 BDL 8 0.005 N/A
2B 7.5 0 1 0 7.70 31 BDL 8 0.005 N/A
2C 7.5 0 1 0 7.61 29 BDL 8 0.005 N/A
3A 7.5 0 4 0 7.36 29 5 4 0.010 N/A
3B 7.5 0 4 0 7.22 30 4 4 0.007 N/A
3C 7.5 0 4 0 7.22 28 7 4 0.016 N/A
4A 8.5 0 0 0 NM 30 32 175 0.066 0.066
4B 8.5 0 0 0 NM 30 15 175 0.030 0.030
4C 8.5 0 0 0 NM 30 12 175 0.025 0.025
5A 8.5 0 1 0 NM 30 13 6 0.027 N/A
5B 8.5 0 1 0 NM 30 6 6 0.013 N/A
5C 8.5 0 1 0 NM 30 4 6 0.009 N/A
6A 8.5 0 4 0 8.27 30 17 2 0.035 N/A
6B 8.5 0 4 0 8.20 29 12 2 0.027 N/A
6C 8.5 0 4 0 8.20 29 7 2 0.015 N/A
7A 10 0 0 0 9.46 32 147 600 0.287 0.287
7B 10 0 0 0 9.50 30 87 600 0.181 0.181
7C 10 0 0 0 9.71 32 161 600 0.319 0.319
8A 10 0 1 0 9.99 31 17 40 0.034 N/A
8B 10 0 1 0 10.26 31 31 40 0.064 N/A
(continued)

Steady-state
ID* pH (mg/L as C) (mg/L as P) (mg/L as Cl2) pH† (min) (nM) ‡ (nM) (mol/min·m2) (mol/min·m2)§
9A 10 0 4 0 NM 30 98 18 0.204 N/A
9B 10 0 4 0 NM 30 20 18 0.041 N/A
9C 10 0 4 0 NM 30 11 18 0.022 N/A
10A 7.5 50 0 0 NM 30 249 1057 0.516 0.516

10B 7.5 50 0 0 NM 33 118 1057 0.223 0.223
11A 7.5 50 1 0 NM 30 25 99 0.052 0.052
11B 7.5 50 1 0 NM 30 23 99 0.048 0.048
12A 7.5 0 0 2 6.22 30 114 220 0.241 0.241
12B 7.5 0 0 2 6.44 30 105 220 0.217 0.217
13A 7.5 0 1 2 6.79 31 BDL 8 0.006 N/A

13B 7.5 0 1 2 6.61 30 BDL 8 0.009 N/A
14A 7.5 50 0 2 7.69 31 37 1057 0.075 0.075
14B 7.5 50 0 2 7.77 30 4 1057 0.008 0.008
15A 7.5 50 1 2 8.00 32 3 99 0.006 0.006
15B 7.5 50 1 2 7.98 31 4 99 0.007 0.007
16A 8.5 50 0 0 NM 30 321 1243 0.677 0.677
16B 8.5 50 0 0 NM 31 206 1243 0.422 0.422
17A 8.5 50 1 0 NM 30 45 128 0.095 0.095
17B 8.5 50 1 0 NM 30 38 128 0.079 0.079
18A 8.5 0 0 2 6.26 34 176 175 0.324 N/A
18B 8.5 0 0 2 5.77 32 133 175 0.260 N/A
19A 8.5 0 1 2 7.54 30 BDL 6 0.005 N/A
19B 8.5 0 1 2 7.56 32 BDL 6 0.008 N/A
(continued)


Steady-state
ID* pH (mg/L as C) (mg/L as P) (mg/L as Cl2) pH† (min) (nM) ‡ (nM) (mol/min·m2) (mol/min·m2)§
20A 8.5 50 0 2 8.27 31 35 1243 0.072 0.072
20B 8.5 50 0 2 8.42 31 59 1243 0.119 0.119
21A 8.5 50 1 2 8.45 31 39 128 0.079 0.079

21B 8.5 50 1 2 8.51 30 39 128 0.081 0.081
‡BDL = below detection limit of 0.5 µg/L (2.4 nM)
§Rate constants were calculated from the measured dissoluton rate and Equation 3.2. N/A indicates that determation of a rate constant was not applicable
because the steady-state effluent concentrations were below detection limits or were not sufficiently below (50 nM difference) predicted equilibrium
concentrations.

0.30
P = 0 mg/L
Lead Releae Rate x 10 (mol/min•m )

2
P = 1 mg/L
0.25 P = 4 mg/L
0.20
9
0.15
0.10
0.05
0.00
7.5 8.5 10
pH
Figure 3.32 Dissolution rates of hydroxylpyromorphite as a function of pH and orthophos-

phate concentration
used to determine dissolution rate constants to determine whether the rate constants followed any
systematic trends with pH or the concentration of the carbonate.
Effect of pH and Orthophosphate
As discussed in Chapter 2, the equilibrium solubility of hydroxylpyromorphite is directly

related to pH and the concentration of orthophosphate. Therefore, it was expected that these two
parameters would have greater effects on hydroxylpyromorphite dissolution rates and steady-state
concentrations (Css) than would the dissolved inorganic carbon (DIC) and chloramine (CA) con-
centrations. Nine chemical conditions were investigated in triplicate (with the exception of pH 10,
1mg P/L which was in duplicate) for varying orthophosphate concentrations and pH.
Dissolution rates. Experiments were conducted at a pH of 7.5, 8.5, and 10 (typical of
drinking water systems) and with orthophosphate concentrations of 0, 1, and 4 mg P/L. Influent
solutions were buffered to maintain the pH and had an ionic strength of 0.01 M by addition of
sodium nitrate. When no orthophosphate was present in the influent solutions, the dissolution rates
were lowest at a pH of 8.5. Rates were about five times higher at pH 7.5 and pH 10 (Figure 3.32).
For conditions with no added orthophosphate or DIC, the dissolution rates of hydroxyl-
pyromorphite were about two orders of magnitude lower than those of hyderocerussite. The rates
were comparable to those of plattnerite at pH 8.5 and 10.0 and slightly higher at pH 7.5.
For a given pH the addition of 1 mg P/L of orthophosphate decreased the dissolution rate.
Further increases in orthophosphate concentration to 4 mg P/L did not significantly change the
rate. Steady-state concentrations when 1 mg P/L was present were low enough to be below the

-3
1 mg P/L
4 mg P/L
-4
1 mg P/L + 50 mgC/L DIC
1 mg P/L +
-5 50 mg C/L DIC
Log[Pb]diss (M)
-6
1 mg P/L
-7 15 ug/L Pb
-8 4 mg P/L
-9
4 5 6 7 8 9 10 11
pH
Figure 3.33 Steady-state effluent and predicted equilibrium concentrations for hydroxyl-
pyromorphite dissolution. Concentrations are presented on a logarithmic scale. Points are
effluent concentrations and the lines are predicted equilibrium concentrations.
detection limit (BDL) of the analytical equipment (ICP-MS). The local minimum in the rate seen
at a pH of 8.5 in the absence of phosphate was not observed when orthophosphate was present.
As the presence of 1 mg P/L is nearly as effective as 4 mg P/L in lowering the steady-state
dissolved lead concentrations, the use of higher concentrations of orthophosphate for lead corro-
sion control is unnecessary. The use of lower concentrations of orthophosphate has the benefits of
lowering the chemical costs associated with corrosion control, minimizing biofilm growth within
distribution systems, and reducing the need for phosphate removal from wastewater. The effect of
adding 1 mg P/L on steady-state concentrations is more pronounced than changing the pH.
Equilibrium versus kinetic control of dissolved lead concentrations. When only ortho-
phosphate was added to the influent, the steady-state effluent lead concentrations were very close
to the predicted solubility of hydroxylpyromorphite (Figure 3.33). The predictions are shown for
the experiments with 1 mg P/L and 4 mg P/L orthophosphate added to the influent because these
conditions allow direct determination of the concentration of PO43– necessary to predict equilib-
rium solubility. Steady-state concentrations were also similar to predicted equilibrium concentra-
tions when no orthophosphate was included in the influent, but these predictions are subject to
greater uncertainty because the only source of the orthophosphate is the dissolution of the solid
itself and slight non-stoichiometry of the hydroxylpyromorphite can have a large effect on the
predicted equilibrium concentration.
As effluent steady-state concentrations are not significantly different than equilibrium con-
centrations, the reactors reached equilibrium on time-scales faster than those of water transport
and dissolved lead concentrations are dependent on thermodynamics and not reaction kinetics.
Dissolution is sufficiently fast that equilibrium is reached within the 30 minute residence of the

0.60

0.55 DIC = 0 mg/L
2
0.50 DIC = 50 mg/L
0.45
0.40
9
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
pH 7.5, P = 0 pH 7.5, P = 1 pH 8.5, P = 0 pH 8.5, P = 1
Figure 3.34 Effect of dissolved inorganic carbon on hydroxylpyromorphite dissolution rates.

Influent conditions are identified with respect to pH and orthophosphate concentration
(denoted by P and with units of mg P/L).
reactor, but the solid is sufficiently insoluble that low dissolved lead concentrations are still main-
tained. The implication of this observation for managing lead concentrations is that dissolution of
hydroxylpyromorphite will probably release dissolved lead quickly to low concentrations that are
near equilibrium solubility and below the action level if dissolved orthophosphate has been added.
Effect of Dissolved Inorganic Carbon
The addition of dissolved inorganic carbon led to an increase in the hydroxylpyromorphite

dissolution rate when other variables were held constant (Figure 3.34). Rates for conditions with
both DIC and orthophosphate were significantly higher than the rates when only orthophosphate
was present. When both DIC and orthophosphate were present, dissolution rates were lower than
when only DIC was present.
Whereas the addition of orthophosphate was expected to have an inhibiting effect on
hydroxylpyromorphite dissolution and to lower steady-state concentrations of dissolved lead, the
addition of carbonate was expected to raise the steady-state concentration of lead by formation of
soluble lead-carbonate complexes. The greatest enhancement induced by DIC addition was at pH
8.5 with no orthophosphate addition, which is consistent with enhancement caused by the forma-
tion of lead(II)-carbonate complexes because the concentration of CO32– is higher at pH 8.5 than
at 7.5 for the same DIC condition.

0.60

Chloramine = 0 mg/L
2
0.55
Chloramine = 2 mg/L
0.50
0.45
0.40
9
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
pH 7.5, DIC = 50, pH 7.5, DIC = 50, pH 8.5, DIC = 50, pH 8.5, DIC = 50,
P =0 P =1 P =0 P =1
Figure 3.35 Effect of monochloramine on hydroxylpyromorphite dissolution rates. The pH

and concentrations of DIC (in mg C/L) and orthophosphate (in mg P/L) are noted.
Equilibrium versus kinetic control of dissolved lead concentrations. Equilibrium con-

centrations for experiments with DIC were predicted and compared with the steady-state effluent
concentrations to determine whether dissolved lead concentrations were controlled by equilibrium
or by kinetics. When orthophosphate and DIC were both present, steady-state concentrations were
significantly lower than predicted equilibrium concentrations, which indicates that orthophosphate
does inhibit the dissolution rate of hydroxylpyromorphite and does not just affect dissolved lead
concentrations by lowering the equilibrium value (Figure 3.33).
Effect of Chloramines
The effect of chloramines on hydroxylpyromorphite dissolution was examined by conduct-

ing experiments with influents containing 2 mg/L chloramines as Cl2. Dissolution rates were lower
in the presence than in the absence of chloramines for experiments that had otherwise identical
influent compositions (Figure 3.35). These included experiments at both pH 7.5 and 8.5 and in
the absence and presence of added orthophosphate. Figure 3.35 only presents experiments with
50 mg C/L DIC included in the influent, because these experiments had stable pH values over the
duration of the experiments. Because the chloramines reacted with the pH buffers used at other
conditions, the pH decreased from the target value for influent solutions that did not have the
buffering provided by the DIC. However, the experiments conducted in the absence of DIC are
also consistent with the presence of chloramines decreasing the dissolution rate. Measurements of

10000
1000
Dissolved Lead (nM)
100
10
0
5.5 6 6.5 7 7.5 8 8.5 9 9.5 10
pH
Figure 3.36 Comparison of steady-state effluent (open squares) and predicted equilibrium
lead concentrations for dissolution of hydroxylpyromorphite in the presence of 2 mg Cl2/L of
chloramines and 1 mg P/L of orthophosphate. The lines indicate the predicted equilibrium
concentrations for hydroxylpyromorphite (solid) and chloropyromorphite (dashed). Mea-
sured concentrations after a 24 hour equilibration are shown as triangles.
chloramine concentrations before and after experiments showed that chloramine concentrations
remained stable as final concentrations were within 90% of the initial concentrations.
The presence of monochloramine probably inhibited dissolution by allowing for the for-
mation of chloropyromorphite (Pb5(PO4)3Cl), which is a less soluble lead-phosphate solid than
hydroxylpyromorphite and that contains chloride in place of hydroxide. In order for chloropyro-
morphite to precipitate there must be chloride in the system. Approximately 0.035 mM of chloride
was added in the form of NH4Cl to influents for chloramine formation. This concentration of
chloride could be sufficient to convert all of the hydroxylpyromorphite in the reactor to chloropy-
romorphite over the course of the experiment. In this manner chloramine itself is not the species
inhibiting dissolution, but rather chloride is the dissolution-inhibiting species. The steady-state
effluent concentrations for influents with chloramines but without DIC are between the predicted
equilibrium concentrations of chloropyromorphite and hydroxylpyromorphite (Figure 3.36), which
is consistent with partial formation of chloropyro-morphite.
Relative Impacts of DIC, Orthophosphate, and Chloramine
The combined impacts of DIC, orthophosphate, and chloramines on hydroxylpyromorphite

dissolution were assessed at pH 7.5 and 8.5 by looking at the trends as different constituents are
added (Figure 3.37). In going from a pure influent to one with DIC, the dissolution rate increases.
The subsequent addition of orthophosphate significantly decreased the dissolution rate and miti-
gated the enhancing effect of the DIC. The addition of chloramines then acted to further decrease
the dissolution rate, with the effect being more pronounced at pH 7.5 than at pH 8.5.

0.60
DIC= 0 mg/L, P = 0 mg/L, CA = 0 mg/L

0.55
2 DIC= 50 mg/L, P = 0 mg/L, CA = 0 mg/L
0.50
0.45
0.40
9
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
7.5 8.5
pH
Figure 3.37 Effects of combinations of water chemistry parameters on hydroxylpyromor-

phite dissolution rates
Post-Reaction Characterization of Solids
For experiments with only orthophosphate added to the influent, the solids recovered from
the reactors after completion of dissolution experiments showed no signs of a change in the hydrox-
ylpyromorphite solid and no precipitation of other lead solids. X-ray diffraction (XRD) patterns
were the same for unreacted hydroxylpyromorphite, recovered hydroxylpyromorphite at different
pH values, and recovered hydroxylpyromorphite at different concentrations of orthophosphate
(Figure 3.38). Solids recovered from experiments with combinations of all three constituents were
also analyzed by XRD to determine if the solid had changed composition. Chloropyromorphite
was the most likely solid to be formed as it is less soluble than hydroxylpyromorphite when chlo-
ride is present. However, its XRD pattern is very similar to the XRD pattern of hydroxylpyromor-
phite and attempts to determine if chloropyromorphite had been formed during the experiments
were inconclusive.
Scanning electron microscopy (SEM) was also performed on these samples to determine
if the morphology of the solid phase changed. Generally, there were no observable differences
between reacted and unreacted solids. However, a few anomalies which may indicate growth of a
different solid on the hydroxylpyromorphite solid surface were found. Scanning electron micro-
graphs are shown in Figures 3.39 through 3.42.

pH 10
pH 7.5,1 mg/L P
Relative Intensity
pH 7.5
Acid Washed Hydroxylpyromorphite
Hydroxylpyromorphite PDF#00-024-0586
15 20 25 30 35 40 45 50 55 60
2θ (°)
Figure 3.38 X-ray diffraction patterns of hydroxylpyromorphite before and after reaction
at different pH and orthophosphate concentrations. The reference pattern for hydroxylpyro-
morphite is also included.
Overall Dependence of Dissolution Rate Constant on Solution Composition
The dissolution rates determined using Equation 3.1 can be used to calculate rate constants
that fit a general rate equation. The dissolution rate may be specific to the particular residence
time and solid loading in an experiment, but the rate constant will be independent of the reactor
configuration and will be only a function of the identify of the reacting solid and the specific water
chemistry. Rate constants were included in the summary table (Table 3.8). Rate constants were
subsequently probed to identify any trends with pH and the carbonate ion.
Specific dissolution rate constants (k), which are only applicable under a certain set of con-
ditions, were calculated using Equation 3.13 for Ω = 1.
k= Rate (3.13)
(1 − 10 SI ) Ω
The saturation index (SI) was calculated using equation 3.14 where the Pb2+ concentration was
calculated from the measured effluent dissolved lead concentration and the PO43– concentration
was calculated from the total dissolved phosphate. The total dissolved orthophosphate for condi-
tions at which no orthophosphate was added was estimated using the stoichiometry of hydroxyl-
pyromorphite of 3 moles of orthophosphate for every 5 moles of lead released. The concentrations
of the free ionic forms, Pb2+ and PO43–, were calculated from the pH and the dissolved lead and
phosphorus concentrations using the reactions included in Appendix B.
6PB 2 +@5 6PO 34 −@3 6H +@− 1

SI = log c IAP e o
Ksp m
= log (3.14)
Ksp, hydroxypyromorphite


Figure 3.39 Hydroxylpyromorphite after reacting with 2 mg Cl2/L of chloramine at pH 8.5

Figure 3.40 Hydroxylpyromorphite after reacting with 1 mg P/L of orthophosphate and
2 mg Cl2/L of chloramine at pH 8.5

Figure 3.41 Hydroxylpyromorphite after reacting with 1 mg P/L of orthophosphate, 50 mg
C/L of dissolved inorganic carbon, and 2 mg Cl2/L of chloramine at pH 8.5. (A) Abnormal
long needle-shaped crystals.
Dissolution rate constants were only determined for experiments with at least a 50 nM difference
between the measured steady-state effluent concentration and the predicted equilibrium concentra-
tion. When the steady-state concentrations are too close to the predicted equilibrium concentration,
the calculation of the rate constant becomes very sensitive to the saturation index, which makes the
rate constant estimates more uncertain. Experiments containing only 1 mg P/L, only 4 mg P/L, and
both 1mg P/L and 2 mg Cl2/L had such low steady-state concentrations that rate constants at these
conditions were not determined.
There is no clear trend in the rate constant with pH (Figure 3.43). Such a trend would be
expected if the rate constant had a form similar to that of Equation 3.15.


Figure 3.42 Hydroxylpyromorphite after reacting with 50 mg C/L of DIC and 2 mg Cl2/L of
chloramine at pH 8.5. (B) Abnormal large flat plate crystal, and (C) abnormal large hexago-
nal crystal.
-8.5
-9.0
logk (mol min m )
-2
-9.5
-1
-10.0
-10.5
DIC = 0 mg/L, P = 0 mg/L

-11.0 DIC = 50 mg/L, P = 0 mg/L
DIC = 50 mg/L, P = 1 mg/L
-11.5
7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0
pH
Figure 3.43 Dissolution rate constant, k (mol·m–2·min–1), of hydroxylpyromorphite as a

function of pH
k = k0 {H +} nH+ {i} ni (3.15)
where, the k0 is the true constant, {H+} is the activity of {H+} (recall that pH = –log{H+}) and nH+ is
the order of the reaction with respect to H+, {i} is the activity of a potentially enhancing or inhibit-
ing species such as carbonate, and ni is the order of the reaction with respect to that species. Such
a relationship to pH has been observed in previous work for dissolution rates of solids, including
lead phosphates, but this relationship has only held at pH values below the range considered in this
project. Rate constants were also examined as potential functions of the carbonate ion concentra-
tion, but any trends were minor and the range of variables considered were not sufficient to firmly
establish any trend.

Summary of Hydroxylpyromorphite Dissolution
The dissolution rates of hydroxylpyromorphite were affected by all four water chemistry
parameters evaluated: pH and concentrations of orthophosphate, DIC, and chloramines. A change
of pH from 7.5 to near pH 8.5 when no orthophosphate, dissolved inorganic carbon, or chlora-
mines were added led to a decrease in the dissolution rate. For comparable conditions with no DIC
or orthophosphate added, the dissolution rate of hydroxylpyromorphite was about two orders of
magnitude less than that of hydrocerussite and comparable to that of plattnerite. The addition of
orthophosphate even more dramatically affected the dissolution rate. Adding 4 mg P/L provided
no advantage over the addition of only 1 mg P/L, indicating that concentrations of 1 mg P/L of
orthophosphate or less are sufficient to control dissolved lead concentrations from pH 7.5 to 10.
This information will allow water utilities to save money from reduced material use and minimize
the growth of biofilms in pipes that orthophosphate (a biological nutrient) can cause.
DIC can increase the rate of hydroxylpyromorphite dissolution, particularly at higher pH.
Increased dissolution rates in the presence of DIC are caused by the complexation of lead(II) by
the carbonate ion. This complexation increases the equilibrium dissolved lead concentration and
may also accelerate the reaction of the surface of the hydroxylpyromorphite. Consequently, high
carbonate alkalinity may actually decrease the overall stability of hydroxylpyromorphite.
Dissolution rates were lower in the presence of chloramines. While the presence of ortho-
phosphate and chloramine may have produced similar effects on hydroxylpyromorphite dis-
solution, the methods through which they inhibit dissolution are different. Orthophosphate can
directly influence dissolved lead concentrations through the solubility relationship of hydroxyl-
pyromorphite. Chloramine is not a species in the solubility relationship of hydroxylpyromorphite.
Measurements of the reacted effluent showed no loss of chloramine during the experiments. The
factor that may cause the decrease in steady-state concentrations when chloramine is present is
chloride ion, which was present in the ammonium chloride solutions used to create chloramine.
Chloride ion is a species in the solubility relationship of chloropyromorphite, which would be less
soluble (and hence the dominant solid species) than hydroxylpyromorphite when chloride was
present.
For most of the experiments conducted when orthophosphate was present and DIC was
absent, the steady-state concentrations were close to predicted equilibrium concentrations. This
indicates that dissolution rates are not important when contact time with hydroxylpyromorphite is
30 minutes or more, as the systems are dominated by equilibrium thermodynamics. Contact times
of greater than 30 minutes are common in lead service lines during times of stagnation. For sce-
narios when water with high DIC concentrations has a contact time with hydroxylpyromorphite
on the order of minutes, then knowledge of dissolution rates may be important in estimating the
dissolved lead concentrations that water consumers may encounter.
While dissolution rates for conditions with orthophosphate present may be more important
than thermodynamics for contact times of a few minutes or less, it should be noted that dissolved
lead concentrations for these scenarios should not be greater than those encountered at equilib-
rium. The dissolved lead concentrations in lead service lines with low DIC concentrations and
added orthophosphate can then be easily predicted using equilibrium solubility relationships.
Dissolution rates for the dissolution of hydroxylpyromorphite were calculated from
steady-state concentrations. These dissolution rates were lower than anticipated from compari-
son to the dissolution rates of similar minerals. Dissolution rates calculated ranged from ~5·10–12
mol m–2 min–1 (~3·10–10 mol m–2 hr–1) for cases where steady-state concentrations were close to

predicted equilibrium concentrations to ~7·10–10 mol m–2 min–1 (~4·10–8 mol m–2 hr–1) for cases
where steady-state concentrations were at least two times less than predicted equilibrium concen-
trations. In comparison, calcium apatites have been observed to have dissolution rates ranging
from 1.6 –2.35·10–7 mol m–2 hr–1 for a pH range of 6–8.5 (Valsami-Jones et al., 1998; Guidry and
Mackenzie, 2003). Chloropyromorphite, which is predicted to be less soluble than hydroxylpyro-
morphite, has a dissolution rate of 1.1·10–7 mol m–2 hr–1 at pH 7 for an undersaturated solution (Xie
and Giammar, 2007). However, the observed dissolution rate of hydroxylpyromorphite at pH 7.5
in an undersaturated solution was ~1.5·10–8 mol m–2 hr–1 (2.5·10–10 mol m–2 min–1), which is a full
order of magnitude less than the rate for chloropyromorphite at similar conditions.
Summary of Rate Measurements and Limitations of Present Study
The dissolution rates of three important lead corrosion products were measured using pure
forms of those solids and systematically examining four important water chemistry parameters.
The exact dissolution rates of those solids have been summarized in the preceding sections. The
dissolution rates were found to be strong functions of pH, DIC concentration, orthophosphate con-
centration, and in some cases also to the presence of chloramines. Two important water chemistry
parameters were not evaluated in this study. The first is the presence of free chlorine, which can
stabilize plattnerite by maintaining a high oxidation-reduction potential and is likely to induce
the transformation of hydrocerussite (Liu et al. 2008; Liu et al. 2009) and possibly hydroxylpyro-
morphite to lead(IV) oxide solids. The dissolution rate and equilibrium solubility of plattnerite is
the subject of a current Water Research Foundation project (Project 4211) that we are conducting.
The second parameter not investigated in this study was the role of natural organic matter (NOM),
which can accelerate the dissolution of plattnerite by acting as a reductant and can also increase the
lead concentrations of lead(II) solids (Korshin et al. 2000; Korshin et al. 2005; Dryer and Korshin
2007; Lin and Valentine 2008a; Lin and Valentine 2009). The effects of NOM on lead release were
a focal area of a Water Research Foundation project led by Valentine (Project 3172).
The dissolution rates of the pure forms of the lead corrosion products provides valuable
information regarding the factors that inhibit or accelerate dissolution, but these results are not
directly applicable to predicting lead concentrations in tap water and caution should be taken to
avoid overinterpreting the results. The flow-through experiments performed in this study have
much larger ratios of surface area to mass than will be encountered in lead service lines. The
complete mixing of the reactors, which was essential to yielding high quality rate measurements,
is also not relevant to lead service lines. Estimations of lead concentrations in actual lead service
lines would require more information regarding the physical structure and exact chemical compo-
sition of the corrosion products. Full information on water chemistry, including free chlorine and
natural organic matter would also be necessary.

CHAPTER 4
RELEASE OF LEAD FROM PIPE SCALES
OVERVIEW OF RESEARCH ON LEAD RELEASE FROM PIPE SCALES
Extending on the work with pure forms of important lead corrosion products, rates of
lead release rates from corrosion products developed as components of scales of lead pipe were
investigated. Two pipe scales were investigated, one on lead pipe of service lines and another
that was developed in the laboratory on new lead pipe. For each of the pipe scales, lead release
was examined during stagnation periods and also following recirculation at different flow veloci-
ties. Lead release was investigated over a range of water compositions (pH, orthophosphate, and
DIC). A conceptual overview of the approach to studying lead release from pipe scales is shown
in Figure 4.1.
The dissolution rates described in Chapter 3 can be evaluated to determine their effective-
ness in predicting lead concentrations in pipe systems. Previous studies showed qualitatively the
effects of stagnation time, flow velocity and water chemistry on lead release from lead pipes. The
stagnation time was found to have a substantial effect on lead release from pipes, with most of
the release occurring within the first 24 hours (Lytle and Schock 2000). Flow velocity can influ-
ence erosion mechanisms of corrosion and can affect the development of pipe scales (Schock
1999). With respect to water chemistry, orthophosphate was demonstrated to inhibit lead release.
However, none of the previous studies used a dissolution rate to predict lead concentrations. The
dissolution model provides a way to quantitatively predict lead release rates from pipe systems by
quantifying the effects of different factors on dissolution rates.
MATERIALS AND METHODS
Lead Pipes With Scales Containing Corrosion Products
Lead Pipe Sections From Lead Service Lines
Many of MWRA’s customer communities have active programs to remove lead service
lines. MWRA arranged for a community to provide ten different sections of lead pipe (Table 4.1).
The lead pipes range from 80 to 102 years old and several lead pipes have no record of the year
of installation. Ozone is currently used as the primary disinfectant and chloramines are added for
disinfectant residual. Since 1997–1998, the MWRA has added sodium carbonate to increase the
pH to 9.5 and reach a target alkalinity of 40 mg/L as CaCO3. These conditions are favorable for
the formation of hydrocerussite (Pb3(CO3)2(OH)2(s)). The current pipe scales most likely reflect
the water chemistry of the past ten years. The specific corrosion products have probably formed
and reformed over time in response to changes in water chemistry with changes in the treatment
process. Prior to the 1930s the water received no treatment, and from the 1930s to around 1997
most were treated only with chloramines with pH adjustment but little alkalinity addition from the
1970s until 1996. In the mid 1990s treatment involved free chlorine followed by chloramines, and
in 1995 the primary disinfectant was changed from free chlorine to ozone.
65

• Following recirculation, examine

Two Pipe Scales lead release after stagnation
• lead service lines from periods of 1, 4, 8, 24, and 48 hours
distribution system • Investigate the effect of recirculation
12-24” flow rate.
• pipe scale developed in
laboratory in presence of
hypochlorite Three Water Compositions
recirculation • moderate pH, high DIC, no orthophosphate

dissolved reservoir • neutral pH, moderate DIC, with orthophosphate
• high pH, moderate DIC, no orthophosphate
phase analysis
Figure 4.1 Experimental pipe reactor system for studying lead release rates from pipe scales
following various stagnation periods and flow velocities
Table 4.1
List of lead pipe supplied by the Massachusetts Water Resources Authority from one of its
customer communities
Pipe Dimensions
Pipe Date Length Inner Diam.
Number Location Year Installed Collected Mineralogy* (in)† (in)
1 East Boston 1 1924 1/23/07 HC, C, Q 12.8 0.56
2 Jamaica Plain 2 1909 1/18/07 HC 13.0 0.66
3 East Boston 2 No Record 1/23/07 HC, C, Q 19.9 0.56
4 Roslindale 1 1927 1/24/07 HC, C 17.1 0.56
5 Roxbury 1 1919 1/17/07 HC 19.1 0.66
6 Dorchester 1 1905 1/26/07 HC, Q 12.3 0.56
7 Roxbury 2 No Record 1/17/07 HC, Q 17.3 0.69
8 East Boston 3 No Record 1/27/07 HC, C, Q 20.9 0.75
9 Mattapan 1927 1/23/07 HC, C, Q 18.0 0.66
10 Jamaica Plain 1 No Record 1/26/07 HC, C, Q 17.6 0.50
* HC = hydrocerussite, C = cerussite, Q = quartz
† Experiments were conducted with 12 in lengths of pipes 1–6.
The condition and shape of the pipe sections were taken into consideration in selecting
six pipes for use in the lead release measurements. Any pipe section that had an elbow or a hole
or that was not at least 12 inches long was not considered. The mineralogy of the remaining pipe
sections were analyzed using X-ray powder diffraction. The six pipe sections with the most hydro-
cerussite were then selected for lead release measurements. A small section (approximately 1 cm2)
of the pipe scale from each pipe was extracted and used to characterize the mineralogy of each
pipe by X-ray powder diffraction. Analyses indicated that each pipe scale contained hydrocerus-
site (Pb3(CO3)2OH(s)), which has a unique characteristic peak at 34.156° (PDF# 00-013-0131)
(Figure 4.2). Cerussite (PbCO3, PDF# 01-070-2062) has a unique characteristic peak at 25.428°
and is present in a few of the pipe scales especially in the scales of Pipes 1, 3 and 4 (Table 4.1 and
Figure 4.2). Scrutinyite and Plattnerite (α-PbO2 and β-PbO2 respectively) were not found in any

Chapter 4: Release of Lead From Pipe Scales | 67
Figure 4.2 Mineralogy of pipe corrosion products of lead pipes removed from the MWRA
distribution system. Reference patterns of constituent phases are included. Numbers corre-
spond to pipe descriptions in Table 4.1.
of the lead pipe scales. This oxidation reduction potential of the water is not high enough to form
Pb(IV) oxides. The disinfectants used by MWRA, ozone and chloramines, would not be expected
to oxidize Pb(II)-containing phases to PbO2. Quartz (SiO2, PDF# 01-073-0834) was found in sev-
eral pipe scales but is believed to be an artifact of the excavation process. Quartz peaks can be seen
in Pipes 1, 3, and 6. Elemental lead (Pb0, PDF# 01-071-6511) has two characteristic peaks (31.34°
and 52.35°) and was seen in several different pipe scales, but it is an artifact of the sampling pro-
cess. A portion of the underlying lead pipe was removed when the pipe scale sample was collected.
Lead Pipes With Scales Developed in the Laboratory
Corrosion products were developed on the interiors of new lead pipe. Thirteen 24-inch
lengths of new lead pipe (Vulcan Lead, Inc.) were reacted with solutions designed to promote the
formation of lead(IV) oxides in pipe scales. The lead pipes were fixed at an inclined angle of 20°
degree in a holding rack. An aqueous solution at pH 10 with 10 mg C/L DIC and 3.5 mg/L free
chlorine was pumped into the lower end of the pipes to avoid air bubbles. After the pipes were
filled, they were kept stagnant for a day and then emptied. The filling procedure was repeated daily

(five days per week) over a period of eight months for four pipes and three months for the other
nine pipes. Effluent samples were normally taken once every week with the exception of the first
months of the first four pipes conditioned, during which time samples were collected daily (five
days per week). The pH and concentrations of residual chlorine and dissolved lead were measured.
At the conclusion of the condition period, one pipe for each conditioning duration was used for
characterization of the pipe scales by XRD and SEM.
Pipe Release Experiments
Lead release experiments were conducted using both the pipe sections from the actual
lead service lines and the pipe sections conditioned in the laboratory for three months. Release
experiments were performed with pipe reactors that allowed controlled flow of water. The reactor
configuration for the experiments with laboratory-conditioned pipes is shown in Figure 4.3. The
pipe sections were filled by pumping water from 500 mL reservoirs with a peristaltic pump. For
recirculation conditions, the outlets of the pipes were connected back to the reservoir.
The configuration used with the pipes from the actual lead service lines is shown in
Figure 4.4. For the experiments with pipe sections provided by MWRA, a 12 inch pipe section
was cut to remove the section used for XRD analysis and to minimize the effect of length when
comparing duplicate conditions. The end of each section of pipe was fitted with plastic barbed
reducers and connected with AquaMend® epoxy such that the epoxy did not come in contact with
the influent water. Flexible tubing was connected to the barbed fittings and passed through the
appropriate barbed reducers and three-way valve. The tubing was passed through a peristaltic
pump that provided flow through the system and connected to a 1.2 L reservoir with a pH probe.
The pH probe provided real-time pH measurements of the recirculation system. The reservoir is
connected to the other side of the pipe section through a series of barbed fittings including another
three-way valve for sampling.
Experiments examined lead release from the pipe reactors for varying (a) water chem-
istry compositions, (b) stagnation times, and (c) flow velocities (Table 4.2). Experiments were
conducted with six sections of pipe prepared in the laboratory and six sections of pipe from the
actual lead service lines. The use of six sections of each pipe enabled duplicate experiments with
three different conditions of water chemistry. The three compositions were selected to represent
three different solution adjustments that can limit lead release (Table 4.3). None of these solutions
contained any free chlorine or chloramines residual. The goal of these experiments was to exam-
ine the release of lead from pipe scales in response to a change in the water chemistry relative to
the condition with which the scales had previously acclimated. The experimental conditions were
sequenced to avoid altering the pipe scales during an experiment. The order in which experiments
were conducted is included in Appendix F. Experiments with high flow velocity were conducted
last to avoid potential problems of physically breaking apart the scales. Duplicate experiments
were sequenced at different times to check the reproducibility of experimental results.
For the evaluation of each stagnation period, fresh solution of the desired chemistry was
recirculated through the pipe section for two hours at a flow velocity of 0.05 or 0.1 m/s (i.e., in a
laminar flow regime). The recirculation experiments were conducted to assess the impact of water
flow on lead release rates; however, the configuration of water recirculating through the same
pipe section is not representative of the flow conditions of lead service line use. The two hours
of recirculation can provide much more extensive lead release than could occur during the much
shorter contact time of flowing water with a pipe section when the water just flows through the

Figure 4.3 System for evaluating lead release from lead pipes with scales developed in the
laboratory. Recirculation was from three 500 mL reservoirs with flow provided by peristaltic
pumps (labeled P2.2, P2, and P1.3).
pipe and is not recirculated. After the recirculation period, the flow to the pipe section was stopped,
the pipe section was sealed, and a sample of the recirculating solution was collected to provide a
measurement of the initial solution conditions. After the prescribed stagnation period, the remain-
ing water in the pipe section was collected. All stagnation times were collected from one pipe in
one continuous period for the actual lead service line sections, but separate experiments were con-
ducted for different stagnation times for the laboratory-conditioned pipes. The pH of the solution
was measured, and the water collected was split into filtered (0.45 μm) and unfiltered samples and
preserved. Filtered samples were analyzed for dissolved lead, and dissolved orthophosphate. The
unfiltered samples were analyzed for total lead; distinguishing between total and dissolved lead
concentrations can provide useful insights regarding corrosion processes (Cantor 2006). Before
starting an experiment to evaluate another stagnation period, the recirculation reservoir was filled
with fresh solution of the desired water chemistry.

pipe section
n
pH probe
peristaltic
Reservoirr pump
Figure 4.4 Pipe reactor for investigating the lead release rates from pipe scales
Table 4.2
Factors evaluated in experiments with pipe reactors
Factor Conditions evaluated Application to investigation
Water chemistry High orthophosphate and neutral pH. Represent three changes to water
Moderate pH and high DIC. chemistry that can mitigate lead release
Moderate DIC and high pH. from corrosion products.
Stagnation time 0, 1, 4, 8, 24, 48 hours Stagnation time controls the dissolved
lead concentration in household
plumbing.
Flow velocity 0, 0.05 or 0.1 m/s, and 0.3 m/s High flow rates can mechanically
degrade pipe scales and release
particulate lead.
For the pipe sections from the actual lead service lines, an additional set of experiments
evaluated the effect of water velocity on lead release to solution. These experiments involved the
recirculation of fresh solution of the desired chemistry through the pipe section for two hours.
Two different flow velocities were analyzed and an additional experiment with no recirculation
was performed as a control. The high velocity of 0.3 m/s (1.0 ft/s) provides flow in a turbulent
regime with a Reynolds number of 5,700 for pipe with a ¾ inch inner diameter. The low veloc-
ity of 0.05 m/s (0.17 ft/s) provides laminar flow. Following the two hour recirculation period, the
water in the pipe reactors was collected as filtered and unfiltered samples and preserved for analy-
sis of dissolved and total lead. Again, these conditions provide information on the effect of water
velocity on lead release, but a recirculation configuration is not representative of the flow through
actual lead service lines and will result in higher lead release than from the once-through flow in
real service lines.
Analysis of Aqueous Composition
Dissolved concentrations of lead and phosphorus were determined by inductively coupled

plasma mass spectrometry (ICP-MS) on an instrument that has an instrument detection limit of
9 ppt (ng/L) and a method detection limit of 50 ppt (ng/L) for lead. Analyses were performed on

Table 4.3
Water compositions evaluated in lead release experiments with pipe sections
Composition
Solution Condition pH DIC (mg C/L) Orthophosphate (mg P/L)
1 Moderate pH and High 8.5 50 0
Dissolved Inorganic Carbon
2 Neutral pH with Orthophosphate 7.5 10 1
3 High pH and Moderate 10.0 10 0
Dissolved Inorganic Carbon
an Agilent 7500ce instrument equipped with an Octopole Reaction System (ORS). The pH of
solutions was measured with a glass pH electrode and pH meter (Accumet). Free chlorine and
chloramine concentrations were measured using the standard DPD colorimetric method (Clesceri
et al. 1999). A scanning ultraviolet/visible spectrophotometer (Perkin-Elmer Lambda 2S) was used
for analysis.
RESULTS AND DISCUSSION
Evolution of Water Composition and Pipe Scales During Conditioning of New Lead Pipe
Water Chemistry During Conditioning Period
The evolution of the water quality during the conditioning of new lead pipes was moni-
tored for eight months for four of the pipes (identified as P1–4) and for three months for the other
nine pipes (identified as P5–13). During the first 10 days of the conditioning period, the total
lead concentration of P1–4 in the water after one day of stagnation initially increased to around
500 μg/L. From 10 days through the remaining period, the total lead concentration dropped to a
lower level and was stable at about 50 μg/L. The pH increased from the initial value of 10 to 11
and then dropped back to 10 during the first two weeks. The residual chlorine concentration after
one day of contact was in the range of 0 to 1 mg/L as Cl2, indicating the consumption of most of
the free chlorine in the filling solution during the oxidation of the lead pipe. The residual chlorine
concentration was close to 0 during the first 10 days. After 10 days, it increased from 0 to 0.5 mg/L
as Cl2 over time and then fluctuated around 0.5 mg/L as Cl2 (Figure 4.5a–c).
For pipes P5–13 the total lead concentration in the water after stagnation times of one day
increased for the first 60 days and then decreased to a stable level around 50 μg/L (Figure 4.6a).
The pH initially increased to approximately 11, and it then decreased back to 10 in several steps
(Figure 4.6b). After 80 days the pH and total lead concentration stabilized. After the one day stag-
nation periods, the free chlorine concentration always decreased from its initial value of 3.5 mg/L
as Cl2 to the range of 0 to 1 mg/L as Cl2 with an average near 0.5 mg/L as Cl2 (Figure 4.6c). This
decrease was caused by the consumption of free chlorine during the oxidation of the lead pipe.
The trends of pH and total lead concentration of pipes P5–13 were generally the same as
in the four pipes that were conditioned for eight months (pipes P1–4). For both sets of pipes the
pH stabilized near 10 and the total lead concentration near 50 μg/L. However, the time of the pH
and total lead concentration decrease was different for these two sets. The total lead concentra-
tion dropped at 10 days for pipes P1–4 and at 60 days for pipes P5–13. Although the free chlorine

a
800
700
600
Lead Conc. (ug/L)
P1
500
P2
400
P3
300
P4
200
100
0
0 50 100 150 200 250
b
11.6
11.4
11.2
11.0
10.8
pH
10.6
10.4
10.2
10.0
9.8
9.6
0 50 100 150 200 250
c 4.0
Residual Chlorine Conc. (mg/L)
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0 50 100 150 200 250
Time (d)
Figure 4.5 Lead concentration (a), pH (b), and residual chlorine concentration (c) in set of
four lead pipes (pipes P1–4) conditioned for 8 months. The initial pH and chlorine concentra-
tion of the filling solution are indicated by the dashed lines.

a
450
400 P5
350 P6
Lead Conc. (ug/L)
300 P7
250 P8
200 P9
150 P10
100 P11
50 P12
0
P13
0 20 40 60 80 100
b 11.2
11.0
10.8
10.6
pH
10.4
10.2
10.0
9.8
0 20 40 60 80 100
c
4.0
Residual Chlorine Conc. (mg/L)
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0 20 40 60 80 100
Time (d)
Figure 4.6 Lead concentration (a), pH (b), and residual chlorine concentration (c) in set of
nine lead pipes (pipes P5–13) conditioned for 3 months. The initial pH and chlorine concen-
tration of the filling solution are indicated by the dashed lines.

Figure 4.7 nXRD patterns of pipes scales formed after conditioning. Pipe P2 had been con-
ditioned for 8 months and pipe P13 for 3 months.
concentrations were largely consumed in both sets, the trends of residual chlorine concentration
were different. The development of the pipe scales at the conditions with free chlorine is important
to the subsequent investigation of the responses of those scales to water chemistry conditions that
do not have free chlorine. The subsequent experiments are useful to understanding the potential
lead release that may occur when a free chlorine residual is no longer provided to the distribution
system.
Evolution of Scales on Lead Pipe
A pipe conditioned for 8 months (pipe P2) and a pipe conditioned for 3 months (pipe P13)
were cut lengthwise to visually observe the scales that developed on the insides of the pipes. The
three-month conditioned pipe (P13) was covered in a white material; the ends contained red areas.
Part of the eight-month conditioned pipe (P2) had red islands in a white and red background. The
reddish materials in ends of the pipes and the islands in pipe P2 indicated the possible presence of
lead(IV) oxides (Lytle and Schock 2005).

Figure 4.8 Particles scraped off from pipe scales of Pipe 2 (a) and Pipe 13 (b)
The materials on the interior pipe wall were gently scraped off with a metal spatula. XRD
analysis of the materials (Figure 4.7) showed that the red islands of pipe P2 contained scrutinyite
and plattnerite, but the major components were still hydrocerussite and litharge (PbO). However,
little or no lead(IV) oxides were identified in other areas of pipe P2, and there was no XRD evi-
dence of any lead(IV) oxides in pipe P13. The facts that the solids were mostly hydrocerussite
and litharge and that lead(IV) oxides were identified in only a few spots suggested that the for-
mation of lead(IV) oxides was kinetically limited. Based on thermodynamic calculations, in the
presence of 3.5 mg/L as Cl2 free chlorine, lead(IV) oxide should be favorable. However, metallic
lead is oxidized first to lead(II) before reaching the lead(IV) state. When lead(II) is the dominant
oxidation state, lead(II) carbonates and oxides can form. As a result, hydrocerussite and litharge
formed before any lead(IV) oxide was produced. Previous research has observed the coexistence
of lead(II) and lead(IV) phases in pipe scales, and there may be layers of different corrosion prod-
ucts in the pipe scales (Schock et al. 2005b). Elemental lead peaks were found in most patterns
because portions of unaltered pipe were scraped off while collecting materials of the pipe scales.
Electron micrographs of the pipe scale materials that were scraped off the pipe surfaces
are shown in Figure 4.8. There are particles with two types of shape in pipes P2 and P13; one is
a platy sheet, and the other is aggregated spherical particles. The hydrocerussite mineral is usu-
ally in a platy shape (Korshin et al. 2005). More platy particles are found in the scales of pipe P2,
while more aggregates of spherical particles exist in the scales of pipe P13. This is consistent with
the percentage of hydrocerussite and litharge in pipes P2 and P13 identified by XRD based on the
heights of peaks from hydrocerussite and litharge.
Pipe cross-sections were also prepared for imaging (Figure 4.9). Before being cut, the
pipes were filled with epoxy to retain the pipe scales. SEM Images were taken to see the layers and
thickness of the pipe scales. There were gaps in the pipe scales and materials of different shapes
identified in different regions, which are indicative of layers. However, no sharp line or strip with a
sudden change of particle shapes was found. Further investigation is needed to determine whether
layers of different crystals exist. The scale of pipe P2 is around 30 μm thick, while the thickness
of the scale in pipe P13 is about 15 μm. The differences in scale thickness are probably caused by
the longer conditioning time of pipe P2.

Figure 4.9 Cross section of pipe P2 surface with development of corrosion products. Unal-
tered lead pipe is visible on the left and the epoxy used to fill the pipe prior to cutting and
polishing is on the right.
Lead Release From Laboratory-Conditioned Pipes
The release of lead from the pipe scales in the laboratory-conditioned pipes was studied as
a function of water chemistry, water flow, and stagnation time. The following sections present the
results of the experiments that investigated the effect of these variables on the release of dissolved
and total lead. Because the pipes were used multiple times for different conditions, the water
chemistry was kept consistent for each pipe and the order of assessing the effects of flow rate and
stagnation time was varied; the schedule of the experiments is included in Appendix F.
Effect of Water Chemistry
The dissolved and lead concentrations were measured following a 2 hour stagnant reac-
tion period of the three different solutions in each pipe. These experiments were conducted twice,
once with a pipe that had not yet experienced recirculating flow and a second time for a differ-
ent pipe that had already experienced recirculating flow. For both cases, Solution 1 (pH = 8.5,
DIC = 50 mg/L) had the highest dissolved and total lead concentrations of the three solutions
(Figure 4.10). Solution 2 (pH = 7.5, DIC = 10 mg/L, orthophosphate = 1 mg P/L) had the lowest
dissolved lead concentrations, although its total lead concentrations were comparable to those of
Solution 3. The orthophosphate in Solution 2 was effective at mitigating lead release from pipe
scales. The dissolved and total lead concentrations for Solution 3 (pH = 10.0, DIC = 10 mg/L)
were intermediate to those of Solutions 1 and 2. It should be noted that the three water chemistry
conditions contained no residual disinfectant of either free chlorine or chloramines; consequently,
the lead concentrations represent the response of the pipe scales that had previously acclimated in
the presence of a disinfectant to conditions in the absence of such disinfectant.
The overall trend in lead concentrations with water chemistry is consistent with the pre-
dicted equilibrium solubility of the solid phases present and with the results of the flow-through
experiments with pure lead-containing phases in Task 2 (Chapter 3). The lowest steady-state efflu-
ent concentrations for all three pure phases studied in Task 2 occurred when the influent solutions
contained orthophosphate. Concentrations at pH 10 and 10 mg C/L DIC were also consistently
lower than those at pH 8.5 and 50 mg C/L DIC. The concentrations in the pipe reactors were higher

35
Dissolved
30 Total
Lead Concentration (ug/L)
25
20
15
10
0
Solution 1 - high DIC Solution 2 - high P Solution 3 - high pH
Figure 4.10 Dissolved and total lead concentrations following 2 hours of reaction with three
different solutions in laboratory-conditioned lead pipes. Results are reported only for the
pipe that had not previously experienced recirculating flow.
than for hydrocerussite flow-through experiments when orthophosphate was present (~15 μg/L
versus ~0.2 μg/L). These higher concentrations can be explained by the depletion of orthophos-
phate during reaction in the pipe reactors, which did not happen in the flow-through experiments
with pure hydrocerussite because orthophosphate was continuously supplied with the influent. For
conditions without orthophosphate, the concentrations in the pipe reactors were lower than steady-
state effluent concentrations for pure hydrocerussite. However, because the pipe scales were mix-
tures of solid phases and not just hydrocerussite, the utility of direct comparisons is limited.
Effect of Water Flow
The effect of water flow was determined by examining the differences in the lead concen-
trations between water in the pipes that had been stagnant for 2 hours and that had been recircu-
lated for 2 hours. In each of Figures 4.11–4.13, the initial condition at the beginning of the 2 hour
period is included because this condition did vary from experiment to experiment.
For all experiments with water flow, the pH converged to the range of 7.5–9 regardless of
the starting pH (Figure 4.11). The high phosphate solution started at pH 7.0 and increased, and the
high pH solution started at 10.0 and decreased. These changes in pH indicate that despite efforts to
maintain the recirculation reservoirs as closed systems, there was some exchange with respect to
the atmosphere. The changes in the pH are consistent with those that would be expected from par-
tial equilibration with atmospheric carbon dioxide. In contrast, there was very little change in the
pH for the pipes that did not have flow because the no flow systems were sealed more completely
and were not able to exchange with the atmosphere.
For Solutions 1 and 3, the solutions without orthophosphate, there was a steep increase in
total lead concentrations for the experiments with flow relative to those with no flow (Figure 4.12).

11.0
Initial
No Flow
10.0 After Recirculation
9.0
pH
8.0
7.0
6.0
Figure 4.11 Change in pH following reaction for 2 hours with and without recirculating flow
200
No Flow -Dissolved
180
After Recirculation - Dissolved
160
No Flow -Total
140 After Recirculation - Total
120
100
80
60
40
20
0
Figure 4.12 Effect of recirculating flow on dissolved and total lead concentrations. Results
are shown for the experiment (out of two replicates) that had not previously experienced
another flow condition.

400
Orthophosphate Conc. (ug/L as P)

350
300
250
200
150
100
50
0
Initial No Flow After Recirculation
Figure 4.13 Effect of flow on orthophosphate concentration. Results are shown for the exper-
iment (out of two replicates) that had not previously experienced another flow condition.
For Solution 2, there was little difference in total lead between flow and no flow conditions.
Dissolved lead shows a more consistent increasing trend than total lead. For all three solutions,
there is a clear increase in the dissolved lead concentrations when there is flow relative to when
there is no flow (Figure 4.12). With recirculation, dissolved lead concentrations may be higher
because the velocity provides mixing and flushes lead away from the pipe walls.
The recirculation of Solution 2 caused a considerable decrease in the orthophosphate con-
centration (Figure 4.13). There was little difference between the orthophosphate concentration in
the initial solution and in solution after the no flow experiments. However, after the experiments
with flow, there was a 75% decrease in both duplicate experiments. The orthophosphate reacts
more completely with flow, and it is consumed by precipitation of lead phosphate solids. It is also
important to note that the initial orthophosphate concentration was between 250 and 300 μg P/L
and not the desired value of 1000 μg P/L.
Effect of Stagnation Time
The stagnation time experiments were conducted with the solutions that remained follow-
ing the two hour recirculation period. The starting condition for the stagnation time experiments
is consequently that after 2 hours of recirculation. The concentration after recirculation is noted
in the figures as time equal to zero (Figure 4.14). For the stagnation time experiments, the way in
which the experiments were conducted is very important in understanding the data. The stagnation
time tests were not all conducted with the same initial solution. The 1, 4, and 8 hour samples were
collected in one experiment, and the 24 and 48 hour samples were collected in a separate experi-
ment that had the same initial solution composition. Because the experiments involved separate
two-hour recirculation periods, the initial concentrations at the time that stagnation began (the
zero hour) were different for the two sets. Therefore there are inherent translation problems that
will be considered when analyzing the figures, but the overall trends (increasing or decreasing)

Solution 1 - high DIC

200
Dissolved Lead (ug/L)

150
100
P6 short
P7 short
50 P6 long
P7 long
Stagnation Time (h)
Solution 2 - orthophosphate
200
P8 short
P9 short
150
P8 long
P9 long
100
50
Stagnation Time (h)
Solution 3 - high pH
200
P11 short
P12 short
150 P11 long
P12 long
100
50
Stagnation Time (h)
Figure 4.14 Dissolved lead concentration over stagnation times following a 2 hour recircula-
tion period

are visible. The recirculation periods provided contact times of flowing water with the pipe scales
that were much longer than would actually be experienced by scales of lead pipes in service with
once-through flow.
The impact of stagnation time on total lead concentrations depended on the solution.
Total lead concentrations consistently decreased with time for Solution 1 for both pipes studied
(Figure 4.15). For Solution 3, the total lead concentrations also decreased with time, but the trend
was not as uniform as for Solution 1. In contrast, total lead concentrations increased with time
for Solution 2. The concentration of lead for Solution 2 had the lowest initial starting point, and
once the orthophosphate was consumed by reaction with the pipe scales, the lead concentrations
may have increased. Solution 1 has the most DIC while Solution 2 has the same amount of DIC as
Solution 3 but it also has orthophosphate.
As already discussed, the pH changed for Solutions 1 and 3 during the 2 hour recircula-
tion period, but during the 48 hours of stagnation the pH was relatively constant. There was also
no considerable trend for the phosphate concentration with stagnation time. The major changes in
phosphate concentration occurred during the recirculation period.
Dissolved Lead Versus Total Lead Concentrations
Nearly all of the lead is in the dissolved form for Solutions 1 and 3 while, on average, only
66% of the lead was in the dissolved phase for Solution 2 (Table 4.4). While the flow values are
relatively similar for all three solutions, the stagnation time values and, most notably, the no flow
values are considerably less for Solution 2. Thus, when there is no flow, lead tends to be mainly
in the dissolved form for Solutions 1 and 3 but only partly in the dissolved phase for Solution 2.
Precipitation of hydroxylpyromorphite in Solution 2 may result in particulate lead that causes
higher total lead concentrations than dissolved concentrations. The concentration at which hydrox-
ylpyromorphite is predicted to form is similar to the concentration in the solutions examined,
which indicates that the solutions may be in equilibrium with hydroxylpyromorphite.
Summary of Lead Release From Laboratory-Conditioned Pipes
Lead release from the laboratory-conditioned pipes was affected by the composition of
the solution in the pipes. Of the three solutions evaluated, the solution with 1 mg P/L orthophos-
phate resulted in the lowest releases of lead to solution. The next most effective solution was the
solution with elevated pH (pH 10), and the highest concentrations were observed for the solution
with high DIC (50 mg C/L) and moderate pH (pH 8.5). These trends are consistent with predicted
equilibrium concentrations. Dissolved and total concentrations were similar for the two solutions
without orthophosphate, but total concentrations were significantly higher for experiments with
orthophosphate present. For these pipes, a lead phosphate solid like hydroxylpyromorphite may
have formed in suspension and contributed to particulate lead loading. Further studies could exam-
ine lead release when a free chlorine or chloramines disinfectant residual is maintained and when
natural organic matter is present, because recent research has illustrated the importance of these
parameters on the stability of lead corrosion products (Boyd et al. 2008; Lin and Valentine 2008b;
Lin and Valentine 2008a; Liu et al. 2008; Lin and Valentine 2009; Liu et al. 2009).
Dissolved and total lead concentrations increased significantly when the high DIC solution
and the solution with orthophosphate were recirculated through the pipes. The total lead increased
with flow more than the dissolved lead did, which indicates that the recirculating flow mobilized

Solution 1 - high DIC

200
Total Lead (ug/L)

150
100
P6 short
P7 short
50 P6 long
P7 long
0
Stagnation Time (h)
Solution 2 - orthophosphate
200
P8 short
Total Lead (ug/L)
150 P9 short
P8 long
P9 long
100
50
Stagnation Time (h)
Solution 3 - high pH
200
P11 short
P12 short
Total Lead (ug/L)
150 P11 long

P12 long
100
50
Stagnation Time (h)
Figure 4.15 Total lead concentration over stagnation times following a 2 hour recirculation
period

Table 4.4
Dissolved lead divided by total lead concentration
Solution 1 Solution 2 Solution 3
No Flow 1.05 0.49 1.09
During Recirculating Flow 0.91 0.84 0.98
During Stagnation 0.93 0.65 0.84
some particulate lead. However, because both dissolved and total lead increased with increasing
flow, the mobilization of particulate is not the only cause of the higher lead concentrations. The
flowing water may have provided greater mixing of the water in the pipes to overcome any limita-
tions of lead release caused by transport of dissolved lead away from the corrosion products on the
pipe surface. It is important to note that while the recirculation experiments provide information on
the effect of flow on lead release, such a recirculation system is not representative of actual water
flow through a lead service line. Recirculation, which allows for several hours of contact of the
pipe scale with flowing water, will result in much greater lead release than would occur in a once-
through configuration of an actual service line.
Lead Release From Pipe Sections of Lead Service Lines
The release of lead from the pipe scales in the pipe sections provided by MWRA was
studied as a function of water chemistry, water flow, and stagnation time. The following sections
present the results of the experiments that investigated the effect of these variables on the release
of dissolved and total lead. Because the pipes were used multiple times for different conditions,
the water chemistry was kept consistent for each pipe and the order of assessing the effects of flow
rate and stagnation time was varied; the schedule of the experiments is included in Appendix F.
Complete results are tabulated in Appendix G, and the following sections present the key findings
in graphical formats.
Effect of Water Chemistry
The amount of lead released to solution was lowest for the solution that contained 1 mg
P/L orthophosphate (Figure 4.16). The greatest release of lead occurred for Solution 3, which had
the high pH of 10 and moderate DIC. These trends were the same for dissolved and total lead
concentrations. Total lead concentrations were consistently higher than dissolved concentrations,
which indicated that a portion of the lead released was in particulate form. Some of the particulate
release may result from the effects of cutting, draining, and shipping the pipe sections used in these
experiments. The greatest differences between total and dissolved lead occurred for the high pH
solution. The results presented in Figure 4.16 are the concentrations following 2 hours of recircula-
tion at the lower flow rate (0.05 m/s). Experiments were conducted for two different pipes, and the
results shown are for the pipe that had not been subjected to any other release experiments prior
to the low velocity recirculation experiment. The results of the replicate experiment conducted on
a similar but not identical pipe after the no-flow condition are presented in Appendix G; the rep-
licate experiments are in very good agreement for Solutions 2 and 3, but the replicate experiment
with Solution 1 had greater variability. High levels of reproducibility were not expected because
the experiments used different pipes that, while they had similar phases in their scales, may have
different thicknesses of products and surface roughness.

400
Dissolved
350 Total
300
250
200
150
100
50
0
Figure 4.16 Dissolved and total lead concentrations following 2 hours of recirculation at
0.05 m/s velocity through lead pipes from MWRA. Results are reported only for the pipe that
had not previously experienced recirculating flow.
The effectiveness of orthophosphate at decreasing the dissolved lead concentrations is con-

sistent with the results of experiments with the laboratory-conditioned pipes and with the results
of flow-through experiments with pure hydrocerussite. However, the concentrations after 2 hours
of recirculation in the pipe reactors were much higher than the steady-state concentrations for the
flow-through experiments with pure hydrocerussite with orthophosphate provided in the influent.
The higher concentrations in the pipe reactors was caused by the loss of orthophosphate from solu-
tion during reaction; the pipe reactors are a closed system in which orthophosphate concentrations
are depleted by reaction with the pipe scales, but the flow-through reactors were open systems
with a continuous supply of orthophosphate provided in the influent. The relationship between
dissolved lead and orthophosphate will be examined again in more detail when the effects of stag-
nation time are discussed.
For conditions without orthophosphate, the dissolved lead concentrations in the pipe reac-
tors can be compared with the steady-state effluent lead concentrations for the experiments con-
ducted with pure hydrocerussite for Task 3. For the solution at pH 8.5 with 50 mg C/L DIC, the
concentration in the pipe reactor of 611 nM (126 μg/L) is comparable to the average steady-state
effluent concentration of 808 nM for pure hydrocerussite and slightly below the predicted equilib-
rium concentration of 944 nM. In contrast, Solution 3 with high pH (pH of 10) and 10 mg C/L DIC
had considerably higher concentrations in the pipe reactors (862 nM or 179 μg/L) than in the efflu-
ents from the flow-through experiments (363 nM) with pure hydrocerussite. These concentrations
even exceed the predicted equilibrium solubility of hydrocerussite at this pH. The unexpectedly
high concentrations for Solution 3 were only observed for the low velocity condition; experiments
with no flow and with the higher velocity recirculation fit the trend with equilibrium concentra-
tions better.

Effect of Flow Velocity
Recirculating flow and the velocity of the water during recirculation affected the amounts
of total and dissolved lead released to solution. The dissolved and total lead concentrations were
comparable for Solutions 1 and 2 for conditions of no flow (after 1 h or stagnation) and of a low
velocity flow after 2 h, and an increase to the high flow velocity for these two solutions resulted
in a significant increase in the dissolved and total lead concentrations (Figure 4.17). The increases
in the dissolved lead concentration may result from greater mixing of the water against the walls
of the pipe and consequently more complete progress of reactions. For total lead concentrations,
a very large increase was observed for Solution 1 with increasing flow rate, with the total lead
exceeding the dissolved lead by almost a factor of 3 for the high flow velocity. The turbulent
conditions of the high flow velocity could have scoured hydrocerussite from the pipe surface and
mobilized lead-containing particles. For the solution with orthophosphate, the dissolved and total
lead concentrations were still very similar at the high flow velocity, and the high velocity did not
mobilize particulate lead. For Solution 3, the concentrations at the highest flow velocity were
noticeably greater than at the conditions of no flow, but the concentrations were actually slightly
lower than those at the lower flow velocity.
As already noted, the low flow velocity results for Solution 3 remain inconsistent with
other results in several regards. For conditions of no water flow, the trend in dissolved lead con-
centrations with water chemistry is slightly different from that with recirculating flow at the lower
flow velocity of 0.05 m/s. Solution 3 (high pH and moderate DIC) had a lower lead concentration
than Solution 1 (moderate pH and high DIC); this same trend is also seen at the high flow velocity
of 0.3 m/s. These results are also more consistent with those from the laboratory-conditioned pipes
and from the experiments with pure hydrocerussite in Task 3, which suggests that the low flow
velocity condition is an outlier for Solution 3.
Effect of Stagnation Time
The majority of lead release occurred within the first 1 hour for pipes contacted with solu-
tions that contained only DIC (Solutions 1 and 3), and an addition hour of recirculation and up to
48 hours of stagnation did not further affect the dissolved lead concentrations (Figures 4.18–19).
These results indicate that dissolution rates of hydrocerussite are sufficiently-fast that dissolved
lead concentrations approached equilibrium values on time-scales much faster than 1 hour. This
conclusion is consistent with that for the flow-through experiments with hydrocerussite that found
that dissolved effluent concentrations could be estimated by equilibrium solubility and not by con-
sidering the dissolution rate. Consequently for stagnation times of 1 hour, and probably shorter,
lead concentrations in pipes with hydrocerussite pipe scales can be effectively estimated by equi-
librium solubility predictions. For conditions of flowing water (without recirculation) that have
very short contact times of the water with the pipe scales, dissolution rates may still need to be
considered to estimate dissolved lead concentrations leaving a section of lead pipe.
The results presented in Figures 4.18–19 also illustrate the mobilization of particulate by
flowing water. The first 2 hours shown in the figures had recirculation at a velocity of 0.05 m/s.
The total lead concentrations were noticeably higher than the dissolved lead concentrations during
this period. Once the flow was stopped after 2 hours, the dissolved and total lead concentrations
became much more similar. The pH of Solution 1 (initial pH of 8.5 and 50 mg C/L DIC) was well-
buffered and remained very constant, but the pH of Solution 3 decreased from 10 to as low as 8.8

a 450
No Flow (1 h)
Diss. Lead Concentration (ug/L)

400
Low Velocity (2 h)
350 High Velocity (2h)
300
250
200
150
100
50
0
b 1200
No Flow (1 h)
Total Lead Concentration (ug/L)
Low Velocity (2 h)
1000
High Velocity (2h)
800
600
400
200
0
Figure 4.17 Effect of recirculating flow on (a) dissolved and (b) total lead concentrations in
MWRA pipe sections. Results are shown for the pipe that was first studied at the low flow
velocity condition.

600
Pipe 2 Total Pb

500 Dissolved Pb
Action Level
400
300
200
100
Time (hr)
600
Pipe 5 Total Pb
Dissolved Pb
500
Action Level
400
300
200
100
Time (hr)
Figure 4.18 Lead concentrations as a function of time for the low flow (0.05 m/s) MWRA pipe
experiments with influent conditions of pH = 8.5 and DIC = 50 mg C/L (Solution 1)

600
Pipe 3 Total Pb
Dissolved Pb
500
Action Level
400
300
200
100
Time (hr)
600
Pipe 6 Total Pb
Dissolved Pb
500
Action Level
400
300
200
100
Time (hr)
Figure 4.19 Lead concentrations as a function of time for the low flow (0.05 m/s) MWRA pipe
experiments with influent conditions of pH = 10 and DIC = 10 mg C/L (Solution 3)

during recirculation and the pH of Solution 2 increased moderately from 7.5 to as high as 8.1. In
actual systems, the continuous inflow of water at a given pH would lead to more stable pH values
and potentially to less lead release for the high pH solution (Solution 3) than was observed in these
experiments. Further, in systems with continuous inflow of water (as opposed to the recirculating
configuration of the experiments in this study) would result in lower net releases of lead because
of the much shorter contact time of a given volume of water with the pipe scale.
For the pipe reactor experiments with orthophosphate present (Solution 2 with 1 mg P/L
initially), the dissolved lead concentrations increased more gradually than for the solutions that
did not contain orthophosphate (Figure 4.20). A stable steady-state concentration is not reached
until 6-10 hours after the start of the 2-hour recirculation period; experiments involved 2 hours of
recirculation followed by 48 hours of stagnation. The slower increase in dissolved lead concentra-
tions was caused by reactions with the orthophosphate. Lead released from the hydrocerussite can
be removed from solution by the precipitation of hydroxylpyromorphite or another lead phosphate
solid. The precipitation of hydroxylpyromorphite offsets the increase in the lead concentration as
long as there is sufficient orthophosphate available. As seen in Figure 4.20, once the orthophos-
phate concentrations are significantly depleted, the dissolved lead concentrations reach a stable
value. Interestingly this stable value remains slightly below the predicted equilibrium concentra-
tion of approximately 150 μg/L for hydrocerussite for conditions similar to those in the pipe (pH ~
8 and 10 mg C/L DIC). Even with the depletion of orthophosphate, orthophosphate addition may
have lasting benefits in mitigating dissolved lead concentrations. If orthophosphate were continu-
ously added to the system, as occurs with water treated with orthophosphate, then a greater con-
version of hydrocerussite to lead phosphates could occur and result in more substantial decreases
in the dissolved lead concentration. Orthophosphate addition is already considered a best practice
at pH values on the low range of those in distribution systems (pH < 8), but chloramines as the
residual disinfectant are less stable in this pH range.
Summary of Lead Release From Pipe Sections From the Lead Service Lines
The influence of water chemistry on lead release from the lead service lines was similar
to its influence on the release from the laboratory-conditioned pipes. Orthophosphate addition
was the most effective of the three approaches for mitigating lead release. Depending on the flow
conditions, the next most effective approach was increasing the pH to 10, although one of the recir-
culating conditions had higher concentrations for increased pH than for the solution that had high
DIC (50 mg C/L at pH 8.5). For the solutions without orthophosphate, the dissolved lead concen-
trations were consistent with measured concentrations from experiments with pure hydrocerussite
in Task 2. With continuous addition of orthophosphate, the solution with orthophosphate would
likely be able to maintain the lowest lead concentrations. As previously noted for the laboratory-
conditioned pipes, further experiments that include a disinfectant residual would provide results at
water chemistry conditions that are more representative of that to which actual lead service lines
are exposed. Additional experiments could also probe the impact of natural organic matter on lead
release.
Recirculating flow increased both dissolved and total lead concentrations. The effect was
minimal in increasing from no flow to a condition with a low velocity that was in the laminar flow
regime, and the effect was much more significant when velocities were high enough to have turbu-
lent flow. As with the laboratory-conditioned lead pipes, the effect of flow on dissolved lead con-
centrations was probably caused by greater mixing of the water with the pipe scales to overcome

200 1500
Total Pb
Pipe 1
Dissolved Pb
Action Level 1200
150
Phosphate (ug P/L)

Total P
Pb Conc. (ug/L)
Dissolved P
900
100
600
50
300
0 0
0 10 20 30 40 50
Time (hr)
200 1500
Pipe 4 Total Pb
Dissolved Pb
Action Level 1200
150
Phosphate (ug P/L)

Total P
Pb Conc. (ug/L)
Dissolved P
900
100
600
50
300
0 0
0 10 20 30 40 50
Time (hr)
Figure 4.20 Lead and orthophosphate concentrations as a function of time for the low flow
(0.05 m/s) MWRA pipe experiments with influent conditions of pH = 7.5, DIC = 10 mg C/L,
and P = 1 mg P/L (Solution 2)

mass transfer limitations to lead release. For the solution with high DIC (Solution 1), the increase
in flow rate also mobilized significant amounts of particulate lead. Recirculation periods of up to
two hours provide much longer contact times of the pipe scales with flowing water than can occur
in actual lead service lines with flowing water. Actual lead service lines have once-through flow
and the contact time as determined by the pipe length and water velocity will be on the order of
second as opposed to hours. Consequently, the recirculation experiments conducted provide infor-
mation on the effect of flowing versus stagnant water on rates of lead release, but the recirculation
is not representative of an actual lead service line.
Stagnation time had little effect on the concentrations of dissolved lead in the pipe sections.
For solutions without orthophosphate, dissolved lead concentrations reached stable values within
1 hour. Consequently, estimations of dissolved lead concentrations in the pipe sections could be
made based only on equilibrium predictions and without regard to dissolution rates. For shorter
contact times between the water and the pipes, as will occur as water flows through the pipes,
the rates may become important in controlling dissolved lead concentrations. When orthophos-
phate was present, the release of lead to solution was slower because the precipitation of a lead
phosphate phase sequestered lead that would otherwise have been released to solution from the
hydrocerussite.

CHAPTER 5
SUMMARY AND CONCLUSIONS
SUMMARY OF PROJECT
The project involved three integrated tasks that focused on the dissolution rates of lead cor-
rosion products. Task 1 was a critical review of literature on the dissolution rates of lead carbonate
and phosphate solids and related phases. The results of this task were presented in Chapter 2, and
the full review is included as Appendix A. The dissolution rates of three important lead corro-
sion products were quantified in Task 2 using continuous-flow stirred tank reactors. The corrosion
products hydrocerussite, hydroxylpyromorphite, and plattnerite were examined over a range of
aqueous compositions that varied with respect to pH and the concentrations of dissolved inorganic
carbon, orthophosphate, and chloramines. The results of this task were presented in Chapter 3.
The chapter included the results of dissolved phase analysis, characterization of the solid phases,
and interpretation of the dissolution process in a reaction-based framework (reactions used are
compiled in Appendix B). Chapter 4 then presented the results of the experiments in Task 2 that
examined the rates of lead release from actual lead pipe sections. Experiments were performed
with pipe sections that had been removed by a water utility after over 80 years of service and with
new lead pipe sections that were conditioned in the laboratory. During the laboratory-conditioning
of the new lead pipe, the evolution of the aqueous phase and solid phases was monitored. Release
rates from the lead pipes were examined as a function of water chemistry, water flow velocity, and
stagnation time.
DISSOLUTION RATES OF LEAD CORROSION PRODUCTS
The dissolution rates of hydrocerussite, plattnerite, and hydroxylpyromorphite were quan-

tified at more than 21 conditions for each solid. In general, the dissolution rates of plattnerite and
hydroxylpyromorphite were comparable and about two orders of magnitude lower than those of
hydrocerussite. Conclusions specific to each solid are presented below, but there are some general
conclusions common to all three solids. Dissolution rates usually increased with decreasing pH.
The complexation of lead(II) by the carbonate ion often led to increased rates with increasing DIC
concentrations. The presence of orthophosphate resulted in very large (more than two orders of
magnitude in some cases) reductions in dissolution rates. Steady-state dissolved lead concentra-
tions in effluents from the reactors were often very close to predicted equilibrium values, which
would obviate the need for rate information in estimating dissolved lead concentrations; however,
for other conditions the rates were clearly important. Rates will be important at shorter contact
times such as those that occur during flow of water through pipe sections.
Dissolution Rate of Hydrocerussite
The dissolution rates of hydrocerussite are strongly influenced by the pH and the concen-
trations of DIC and orthophosphate. Dissolution rates decrease with increasing pH. The influence
of DIC is complex for hydrocerussite because of two concurrent effects that operate in opposite
directions. The carbonate ion present with DIC lowers the equilibrium solubility of hydrocerussite
93

and decreases the dissolution rate, but the complexation of lead(II) by carbonate increases the solu-
bility; consequently, there is a local minimum in the effect of increasing DIC concentration on the
dissolution rate. Orthophosphate lowers the dissolution rate of hydrocerussite by more than two
orders of magnitude. The effect of orthophosphate is associated with the precipitation of hydrox-
ylpyromorphite, a solid whose formation was directly observed with XRD and SEM. Chloramines
did not significantly affect the dissolution rate of hydrocerussite.
The continuous-flow reactors were chosen for the study because of their utility in measur-
ing dissolution rates and not just equilibrium; however, for many conditions with hydrocerussite,
the dissolution reaction had reached equilibrium on time-scales shorter than the one hour residence
time of the flow-through reactors. For these conditions, estimates of dissolved concentrations in
lead service lines can be made based only on equilibrium and without regard to reaction rates.
Kinetics may play a role during conditions of flowing water, but for stagnation times of one hour
or more, equilibrium estimations can provide good estimates of dissolved lead.
Dissolution Rate of Plattnerite
Plattnerite dissolution rates were influenced by all four water chemistry parameters stud-
ied. Dissolution rates increased with decreasing pH. The formation of soluble lead(II)-carbonate
complexes caused higher dissolution rates in the presence of DIC. The observed effect of DIC
helped establish that lead(IV) oxide dissolution proceeds through reduction of lead(IV) to lead(II)
at the surface of the solid phase. The presence of orthophosphate strongly inhibited lead release
during plattnerite dissolution. The effect of orthophosphate is likely caused by precipitation of lead
in hydroxylpyromorphite and also by the adsorption of phosphate to plattnerite to block potential
dissolution sites. Dissolution rates were also lower in the presence of chloramines, and the reasons
for this effect involve unresolved questions regarding interactions of chloramines with lead(IV)
oxides.
Dissolved lead concentrations in the effluents of the reactors were usually controlled by dis-
solution rates of plattnerite and were well below its predicted equilibrium solubility. Consequently,
estimates of dissolved lead concentrations in lead service lines containing plattnerite should con-
sider the rates of the reaction and not just equilibrium. When orthophosphate was present, hydrox-
ylpyromorphite was not observed, but the dissolved concentrations were consistent with the equi-
librium solubility of this solid.
Dissolution Rate of Hydroxylpyromorphite
Hydroxylpyromorphite dissolution was affected by all four water chemistry parameters.

The dissolution rate depended on the pH in a manner that mirrored the influence of pH on the equi-
librium solubility of hydroxylpyromorphite. Dissolution rates were lower at pH 8.5 than at pH 7.5
or 10. The presence of DIC increased the dissolution rate through the formation of soluble lead car-
bonate complexes. Orthophosphate addition consistently lowered the dissolution rate. Dissolution
rates were comparable for 1 mg P/L and 4 mg P/L orthophosphate, and effective control of dis-
solved lead concentration may even be achievable at concentrations below 1 mg P/L. Dissolution
rates were lower in the presence of chloramines, but the cause was most likely from the reaction
with chloride to form the low solubility lead phosphate solid chloropyromorphite.
Dissolved concentrations in the reactor effluents were very close to the predicted equilib-
rium solubility when DIC was absent, but concentrations were less than the predicted solubility

Chapter 5: Summary and Conclusions | 95
when DIC was present. Because most water supplies contain some DIC, dissolved lead concen-
trations in premise plumbing can be controlled by dissolution rates and not just by equilibrium.
Although dissolution rates may be important in controlling dissolved lead concentrations, when
orthophosphate is present even the equilibrium dissolved lead concentrations are below the action
level for lead.
LEAD RELEASE FROM SCALES ON LEAD PIPES
Corrosion Products Present in Pipe Scales
The pipe sections supplied by the MWRA had well-developed pipe scales that were com-
posed almost exclusively of the lead carbonate solids hydrocerussite and cerussite. The formation
of these phases is consistent with the solution chemistry of the water in the MWRA distribution
system. Hydrocerussite was the more common solid, and release rate experiments focused on six
lead pipe sections that predominantly contained hydrocerussite.
New lead pipe sections that were treated with a high pH (pH 10) solution and 3.5 mg/L as
Cl2 free chlorine had aqueous phases and corrosion products that evolved over time. During the
early periods of conditioning, the pH and dissolved lead concentrations increased significantly.
They then stabilized at pH 10 and about 50 μg/L dissolved lead after 10–60 days. The continuing
consumption of free chlorine indicated that corrosion products were still developing through oxi-
dation by free chlorine. The laboratory-conditioned pipes developed corrosion products of hydro-
cerussite and litharge (a lead (II) oxide), and only after 8 months of conditioning were lead(IV)
oxides observed. The development of lead(IV) oxides is a slow process that is kinetically limited,
and lead(II) and lead(IV) solids can likely coexist for long periods.
Lead Release From Pipe Sections
Orthophosphate addition was the most effective strategy for mitigating lead release for
both the laboratory-conditioned lead pipes and those from the lead service lines. The next most
effective approach was increasing the pH to 10 with 10 mg C/L DIC, and the least effective
approach was increasing the DIC to 50 mg C/L at pH 8.5. The effectiveness of orthophosphate
was ultimately limited by the consumption of the orthophosphate during recirculation and stagna-
tion; this limitation will not apply to actual premise plubming with orthophosphate because of the
continuous influx of orthophosphate with the flowing water. For conditions of no flow, the trends
in concentration and the actual concentration were generally consistent with the results of experi-
ments in Task 2 that were conducted with pure forms of the corrosion products. The actual lead
service lines in particular, which had well-developed scales of hydrocerussite, behaved similarly
to the flow-through experiments with pure hydrocerussite.
Actively flowing water consistently increased the dissolved and total lead concentrations in
the pipe. The mixing induced by the water flow probably provided greater contact of the pipe scales
with the water in the pipe and avoided any limitations of transport of dissolved lead away from
the pipe surface. The flowing water also mobilized some particulate lead, although this effect was
minimal at the low recirculation velocity corresponding to the laminar flow regime. Mobilization
of particulate lead was most significant for turbulent flow with the actual lead service line pipe sec-
tions with the solution with high DIC. The recirculation configuration of flow in the experiments

provided much longer contact times, and consequently higher lead concentrations, than will occur
during the once-through flow for pipes in service.
Dissolution rates were sufficiently high in the experiments with lead pipes that equilibrium
concentrations were reached on time-scales of 1 hour or less for most conditions. Subsequent recir-
culation and stagnation time had little effect on dissolved lead concentrations. At least for pipes
with substantial hydrocerussite in the pipe scales, estimates of dissolved lead in pipe sections can
often be made based only on equilibrium predictions and would not need to account for rates of
dissolution. For shorter contact times, such as during active flow of water, dissolution rates may be
important. Stagnation time did affect the dissolved lead concentrations for the lead service line pipe
sections filled with the solution containing orthophosphate. In these experiments, dissolved lead
concentrations increased more slowly because the net release of lead to solution was controlled by
the balance between hydrocerussite dissolution and hydroxylpyromorphite precipitation.

CHAPTER 6
RECOMMENDATIONS TO UTILITIES
IMPORTANCE OF COLLECTING DATA ON WATER AND SOLIDS
The project results highlight the importance of knowing the composition of both the
aqueous phase and solid phases in estimating and managing lead concentrations in tap water.
Information on the water chemistry at various points in distribution systems can be valuable in
predicting regions of systems that are most vulnerable to having elevated lead levels. Such infor-
mation on water chemistry could be gained at the same time as sampling activities for regular
monitoring programs, including those conducted for compliance with the Lead and Copper Rule.
When samples are collected for measurements of lead and copper, samples could also be collected
for measurements of pH, alkalinity, orthophosphate concentration, and the concentration of the
disinfectant. Datasets that include these water chemistry parameters together with lead concentra-
tions can be useful for more definitively determining the factors controlling lead concentrations.
The release rates of lead are strongly affected by the composition of the corrosion products
that are present in scales on lead pipes, but the identity of these scales is not frequently deter-
mined for actual lead service lines. Sampling and analysis of pipe scales as part of lead service
line replacement plans can provide insights into the most likely corrosion products present in the
lead service lines that remain in place. Knowledge of the pipe scales can then guide the selection
of strategies for limiting lead release from the corrosion products. Because these lines contain the
actual corrosion products present at the time of extraction, sections of these lines can be saved and
then used later to study the potential lead release in response to proposed changes in distribution
system water chemistry.
POTENTIAL PROCESSES IMPACTING CORROSION CONTROL
The effectiveness of corrosion control strategies will vary depending on the source water
chemistry and the composition of the pipe scales. For systems with histories of free chlorine use
as the residual disinfectant, lead(IV) oxides such as plattnerite may control the release of dissolved
lead concentrations. Maintaining free chlorine concentrations will likely maintain the stability of
the corrosion products and limit lead release to solution. In the absence of a free chlorine residual,
the lead(IV) oxides will dissolve and their dissolution rates can be affected by the water chemistry.
Dissolution rates will be lower at higher pH values. Intriguingly, increases in carbonate alkalin-
ity that are likely to accompany increases in pH may actually result in greater dissolution rates
because of the mobilization of lead(II) in the form of soluble complexes with carbonate. Net disso-
lution rates can be substantially reduced by the addition of orthophosphate as a corrosion inhibitor.
Many lead service lines are likely to have hydrocerussite as a dominant phase in pipe
scales. The stability of this phase increases with increasing pH. The effect of adding DIC to
increase alkalinity is complex for hydrocerussite, and equilibrium solubility calculations can pro-
vide a good guide in selecting an optimal DIC addition; as with plattnerite, the addition of too
much carbonate alkalinity to systems with hydrocerussite actually increases dissolved lead con-
centrations. Lead service lines with lead concentrations controlled by hydrocerussite are likely
97

to have higher concentrations than other systems. Orthophosphate addition can be an effective
means of mitigating lead concentrations.
Orthophosphate addition promotes the transformation of hydrocerussite into the lower sol-
ubility phase of hydroxylpyromorphite. If dissolved chloride concentrations are sufficiently high,
which they likely are for many source waters, then the less soluble phase chloropyromorphite may
form. The stability of any hydroxylpyromorphite that is formed can be maintained with residual
orthophosphate and the prevention of low pH conditions. Orthophosphate concentrations do not
need to be maintained at very high levels; this work found almost no benefit in increasing the
orthophosphate dose from 1 mg P/L to 4 mg P/L. A minimal dose may be established that mitigates
lead release effectively and minimizes cost and the potential for negative side effects of biological
growth and the need for orthophosphate removal in wastewater treatment.

APPENDIX A
REVIEW OF LEAD CARBONATE, PHOSPHATE, AND OXIDE
DISSOLUTION RATES
LEAD CORROSION PRODUCTS AND DRINKING WATER QUALITY
Contribution of Lead Corrosion Products to Dissolved Lead in Drinking Water
Lead release from pipes and other components of premise plumbing is a serious threat to
public health. The recent observations of extremely high lead levels in Washington D.C. tap water
illustrate the current importance of this problem (Edwards and Dudi 2004; Renner 2004). While
advances in water treatment, supply, and distribution were ranked fourth on the list of 20th Century
engineering achievements compiled by the National Academy of Engineering, management of our
water treatment and supply infrastructure has been identified as a great engineering challenge for
the 21st century, and corrosion control plays a leading role in this challenge (Edwards 2004). The
American Society of Civil Engineers gave drinking water a D– in its 2004 report card for infra-
structure (American Society of Civil Engineers 2005).
the premise plumbing. Lead release can occur within components owned by the utility or in the
plumbing of individual connections. The release of lead from metal pipes and fittings is controlled
by both the water composition and the properties of the pipe surface. Lead may be released directly
from the pipe or from lead-containing corrosion products on the pipe surface. Lead can also be
released from other lead-containing metals, particularly brass.
Information on lead corrosion rates is particularly valuable as water suppliers consider
process changes that affect water chemistry such as switching disinfectant type or dose, adjusting
pH, or adding a corrosion inhibitor. While the formation of lead corrosion products can passivate
the pipe surface and limit further lead release, changes in water chemistry can cause dissolution of
the corrosion products and lead release to solution (AWWA 2005). The increase in lead concentra-
tions in the Washington D.C. Water and Sewer Authority service area occurred following a switch
to chloramination for disinfection, which dramatically demonstrates the lead release that can result
from changes in water chemistry.
Lead corrosion products observed on lead pipes include lead(II) carbonates and phosphates
(Schock 1999) and, more recently, lead(IV) oxide (PbO2(s)) (Edwards and Dudi 2004; Lytle and
Schock 2005). The lead carbonates cerussite (PbCO3(s)) and hydrocerussite (Pb3(CO3)2(OH)2(s))
are formed in systems for which the solution pH and dissolved inorganic carbon are sufficiently
high to be saturated with respect to these solids. Recent studies have directly observed cerussite
and hydrocerussite on lead and brass materials using X-ray diffraction (XRD) (Hozalski et al.
2005), cyclic voltammetry and electrochemical impedance spectroscopy (El-Egamy 1995), and
extended X-ray absorption fine structure spectroscopy (Frenkel and Korshin 1999). In many cases
the corrosion layers are similar to pure phases but are more disordered (Frenkel and Korshin 1999).
Cerussite and hydrocerussite can also form at the surfaces of calcium carbonate solids through a
99

reaction sequence involving lead adsorption, calcium carbonate dissolution, and lead carbonate
re-precipitation (Godelitsas et al. 2003; Rouff et al. 2004).
Lead phosphate solids are only likely to form in systems to which orthophosphate has been
added as a corrosion inhibitor. Hydroxylpyromorphite (Pb5(PO4)3OH(s)) is a commonly observed
lead phosphate corrosion product. For example, a recent study of pipe in Glasgow, Scotland used
XRD and secondary ion mass spectrometry (SIMS) to identify lead phosphate solids as corrosion
products (Davidson et al. 2004). In the presence of chloride, chloropyromorphite (Pb5(PO4)3Cl(s))
can form as a very insoluble lead corrosion product. Chloropyromorphite was recently observed
in corrosion products in lead pipes, but the dissolved lead concentrations were still higher than the
equilibrium solubility of chloropyromorphite (Schock et al. 2005a).
Lead(IV) oxides have recently been observed as a component of the scales on lead pipe
from several communities (Lytle and Schock 2005). In a study of pipe sections removed from
34 different communities that represented a range of water chemistries, 26% had samples that
included lead(IV) oxides. Investigation of these pipe scales as a function of depth observed three
different modes of lead(IV) oxide occurrence: uniform scales, patchy coverage with coexistence
with hydrocerussite, and a layered structure with plattnerite on top of cerussite (Schock et al.
2005b). Although lead is not stable in the +IV oxidation state in oxygenated water, this oxidation
state is stable at the high oxidation-reduction potential (ORP) provided by free chlorine. Lead(IV)
precipitates in very insoluble PbO2(s) phases that can maintain low dissolved lead concentrations.
Because PbO2 is only stable at high ORP, the decrease in the ORP that occurs when chloramines
are used instead of free chlorine causes the PbO2 to breakdown and release lead back to solution
(Vasquez et al. 2006). Laboratory and pilot-scale studies have observed higher dissolved lead lev-
els with chloramines than with free chlorine, which can be explained by the differences in solubil-
ity of Pb(IV) and Pb(II) phases (Edwards and Dudi 2004; Vasquez et al. 2006). The formation of
PbO2 was observed upon reaction of dissolved lead with sodium hypochlorite, and Pb(II) carbon-
ate phases were observed as intermediate products that can also co-exist with PbO2. The formation
of PbO2 is reversible, with PbO2 disappearing when a residual hypochlorite concentration is not
maintained. Two polymorphs of PbO2 are formed: scrutinyite (α-PbO2) at high pH and plattnerite
(β-PbO2) at neutral to low pH (Lytle and Schock 2005). Hydrocerussite can be transformed to
scrutinyite following a lag period when reacted with free chlorine (Liu et al. 2006).
Corrosion Control Strategies to Limit Dissolution of Corrosion Products
Many water utilities must implement a corrosion control strategy to minimize dissolved
lead concentrations in tap water. Recommended strategies for controlling lead corrosion include
pH and/or alkalinity adjustment and addition of orthophosphate (Edwards et al. 1999; Schock
1999). The general strategy of corrosion control is to promote the formation of low solubility lead-
containing solids that passivate the surface. An improved understanding of the effects of water
chemistry on lead release rates can be used to identify strategies that inhibit dissolution and mini-
mize lead release for specific systems.
Based on a survey of utilities, large decreases in soluble lead levels are observed with
increasing alkalinity. Lead concentrations were highest in systems with alkalinity less than 15 mg/L
as CaCO3, and the largest decreases in dissolved lead occurred as the alkalinity increased from less
than 30 mg/L to greater than 30 mg/L as CaCO3 (Edwards et al. 1999). The effect of increasing

Appendix A: Review of Lead Carbonate, Phosphate, and Oxide Dissolution Rates | 101
alkalinity is consistent with lead concentrations being determined by the solubility of lead carbon-
ate (cerussite and hydrocerussite) solids. The stability of these carbonate solids is determined by
the activity of the carbonate ion, which is determined by both the DIC (or alternatively alkalinity)
and pH. Consequently, increasing pH without also increasing carbonate alkalinity will not be as
effective. In fact, without sufficient alkalinity (less than 20–30 mg/L as CaCO3), increasing pH can
actually exacerbate lead release (Dodrill and Edwards 1995).
Phosphate inhibitors can effectively mitigate lead concentrations. The form in which the
phosphate is added is vitally important. Orthophosphate and associated forms (e.g., zinc ortho-
phosphate) are the most effective inhibitors (Edwards et al. 1999; Schock 1999; Churchill et al.
2000; Cantor et al. 2003). Orthophosphate inhibitors can result in the precipitation of insoluble
lead phosphate solids that limit lead release to solution. In contrast, polyphosphates form dissolved
complexes with lead and can increase dissolved lead concentrations (Dodrill and Edwards 1995;
Edwards and McNeill 2002; McNeill and Edwards 2002; Schock et al. 2005a; Cantor 2006). The
benefits of orthophosphate as a corrosion inhibitor are most significant for systems with relatively
low alkalinity or pH, conditions at which lead concentrations are not controlled by lead carbonate
solids. In a survey of utilities, the most pronounced benefits of orthophosphate addition were seen
in waters with less than 30 mg/L alkalinity as CaCO3 (Dodrill and Edwards 1995). Recommended
orthophosphate doses can range from less than 1 mg/L to as much as 4.5 mg/L as P (Edwards et
al. 1999).
Although orthophosphate addition can effectively control lead concentrations, it can have
detrimental effects such as the promotion of bacterial growth within the distribution system. A
recent pipe loop study using orthophosphate observed a significant increase in bacteria as mea-
sured by heterotrophic plate counts (Hozalski et al. 2005). A survey of 2500 utilities that had
exceeded lead or copper action limits indicated that most utilities are aware of the potential draw-
backs of phosphate addition; however, utilities still select inhibitors based on only limited informa-
tion (McNeill and Edwards 2002).
Purpose of This Literature Review
While several studies have evaluated the equilibrium solubility of lead corrosion prod-
ucts, there have been few studies of dissolution and transformation rates, and fewer still that were
designed to develop general rate laws for dissolution. Information about the equilibrium solubility
and the dissolution rates of lead-containing solids is distributed in the literature of fields including
corrosion science and geochemistry as well as water treatment and supply. This review compiles
and synthesizes information from the literature to make the results and implications of those previ-
ous studies more accessible to the water supply community.
EQUILIBRIUM SOLUBILITY OF LEAD CORROSION PRODUCTS
Maximum dissolved lead concentrations are determined by the equilibrium solubility of

the corrosion products. Equilibrium solubility has been thoroughly reviewed by Schock (Schock
1999), and only a brief overview is provided here.

Table A.1
Equilibrium constants and reactions for dissolved lead species included in the
calculation of total dissolved lead concentrations
# Reaction Log K*
1 Pb2+ + H2O PbOH+ + H+ –7.60
2 Pb2+ + 2H2O → Pb(OH)20 + 2H+ –17.09
3 Pb2+ + 3H2O → Pb(OH)3– + 3H+ –28.09
4 Pb2+ + 4H2O → Pb(OH)42– + 4H+ –39.70
5 Pb2+ + CO32– → PbCO30 6.48
6 Pb2+ + 2CO32– → Pb(CO3)22– 9.38
7 Pb2+ + CO32– + H+ → PbHCO3+ 13.20
8 Pb4+ + 3H2O → PbO32– + 6H+ –23.04
9 Pb4+ + 4H2O → PbO44– + 8H+ –63.80
*Equilibrium constants for Pb(II) species are from the default database of MINEQL+ version 4.5 (Schecher and
McAvoy 1998), and those for the Pb(IV) complexes were calculated from published free energies of formation
(Pourbaix 1974).
Chemical Reactions That Determine Solubility
Lead(II) Carbonate and Phosphate Solids
The total dissolved lead concentration is the sum of the concentrations of the free lead ion
(Pb2+), lead hydrolysis species (e.g., PbOH+), complexes with inorganic anions (e.g., PbCO3(aq)),
and complexes with organic ligands (e.g., humic substances in natural organic matter) as in
Equation A.1.
Pbdiss = [Pb2+] + OH-complexes + Pb-CO3 complexes + Pb-organic complexes (A.1)
Natural organic matter can enhance the solubility of lead in equilibrium with lead solid phases, and
it can also inhibit the formation of those phases (Korshin et al. 2000; Korshin et al. 2005). The dis-
tribution of dissolved lead among these species is governed by the reactions listed in Table A.1 and
is illustrated graphically in Figure A.1. Additional polynuclear lead complexes (e.g., Pb2(OH)3+)
can form, but they are negligible species at the relatively low lead concentrations relevant to drink-
ing water. In these reactions and those involving solids, it is important to note the species used to
write the reaction. In this document all reactions are written in terms of H+ and not OH–.
When solids are present, the concentration of the free lead ion can be calculated from the
equilibrium solubility product for the specific solid. The solubility of the lead carbonate solids
cerussite and hydrocerussite are controlled by their equilibrium solubility products (Ksp) given as
Equations A.2–3a for the dissolution reactions A.2–3b.
[Pb 2 +] 3 [CO 23 −] 2
Ksp, hydrocerussite = = 10 − 18.77 (A.2a)
[H +] 2

100
2+
Pb
90
Percent of Dissolved Lead(II) 80 PbCO3(aq)

70
60 Pb(OH)2(aq)
50
40
Pb(CO3)22-
30 Pb(OH)3-
PbHCO3+ +
20 PbOH
10
0 Pb(OH)42-
4 5 6 7 8 9 10 11
pH
Figure A.1 Distribution of dissolved lead species as a function of pH for a solution contain-
ing 15 μg/L of lead and 50 mg/L dissolved inorganic carbon. Calculations were performed
without considering the precipitation of solid phases.
Pb3(CO3)2(OH)2(s) + 2H+ = 3Pb2+ + 2CO32– + 2H2O (A.2b)
Ksp, cerussite = [Pb 2 +] [CO 23 −] = 10 − 13.13 (A.3a)
PbCO3(s) = Pb2+ + CO32– (A.3b)
Similarly several lead phosphate solids, including hydroxylpyromorphite and chloropyromorphite,

can potentially form on lead pipes and are described by Equations A.4–5.
[Pb 2 +] 5 [PO 34 −] 3
Ksp, hydroxylpyromorphite = = 10 − 62.79 (A.4a)
[H +]
Pb5(PO4)3OH(s) + H+ = 5Pb2+ + 3PO43– + H2O (A.4b)
Ksp, chloropyromorphite = [Pb 2 +] 5 [PO 34 −] 3 [Cl −] = 10 − 84.43 (A.5a)
Pb5(PO4)3Cl(s) = 5Pb2+ + 3PO43– + Cl– (A.5b)
The pH, as expressed by [H+], carbonate ion concentrate ion [CO32–], and orthophosphate concen-
tration [PO43–] determine the equilibrium concentration of the free lead cation [Pb2+].

Both the carbonate and phosphate ion concentrations are strongly pH-dependent. The car-
bonate ion concentration can be determined from the pH and dissolved inorganic carbon concen-
tration (DIC) in Equation A.6.
DIC = [H2CO3*] + [HCO3–] + [CO32–] (A.6)
nate the alkalinity as in Equation A.7, then DIC can be calculated from pH and alkalinity.
Carbonate Alkalinity = –[H+] + [OH–] + [HCO3–] + 2[CO32–] (A.7)
However, as Schock makes clear in his review (Schock 1999), the carbonate ion concentra-
tion is the property that truly determines lead carbonate solubility, and that concentration is most
appropriately determined by the relationship between pH and DIC. The phosphate ion concentra-
tion can similarly be determined by the pH and total dissolved orthophosphate concentration in
Equation A.8.
Dissolved Orthophosphate = [H3PO4] + [H2PO4–] + [HPO42–] + [PO43–] (A.8)
Lead(IV) Oxide Solids
In the presence of strong oxidants, HOCl/OCl– in particular, lead can be oxidized from the
+II oxidation state to the +IV oxidation state. The reduction half-reaction is given as Equation A.9.
Lead(IV) forms a low solubility oxide (PbO2) following the reaction and solubility product given
as Equation A.10a–b.
Pb4+ + 2e– = Pb2+, EH° = 0.845 V (A.9)
[Pb 4 + ]
Ksp, scrutinyite = = 10 − 8 . 26 (A.10a)
[H + ] 4
PbO2(s) + 4H+ = Pb4+ + 2H2O (A.10b)
A complete reaction for the dissolution of PbO2(s) coupled with HOCl is given as Equation A.11.
PbO2(s) + Cl– + 3H+ = Pb2+ + HOCl + H2O (A.11)
For this case, the divalent free lead ion concentration is determined by the oxidation-reduction
potential (ORP) set by the concentrations of chloride and hypochlorous acid and by the pH. While
free chlorine is sufficiently oxidizing to promote the formation of lead(IV), monochloramine can
oxidize metallic lead (i.e.., lead(0)) to lead(II) but not to lead(IV) (Switzer et al. 2006). Direct elec-
trochemical measurement of the ORP values of monochloramine and free chlorine have recently
confirmed the inability of monochloramine to stabilize lead(IV) oxides (Rajasekharan et al. 2007).

a) -1 b) -1
Hydrocerrusite + Pb(OH) 2
Cerrusite
Cerrusite
Hydrocerrusite
-2 -2
Hydrocerrusite + Pb(OH) 2
Hydrocerrusite
Cerrusite
Pb(OH)2
Hydrocerrusite
+
-3 -3
Pb(OH) 2
Log[Pb]diss (M)
Log[Pb]diss (M)
-4 -4
-5 -5
-6 -6
-7 -7
Dissolved Lead Dissolved Lead

-8 -8
-9 -9
4 5 6 7 8 9 10 11 4 5 6 7 8 9 10 11
pH pH
Figure A.2 Dissolved lead concentration and solid phases calculated to be present for a sys-
tem initially containing 1 g/L hydrocerussite. Plots shown for (a) 10 mg/L or (b) 50 mg/L of
dissolved inorganic carbon.
Estimated Lead Solubility as a Function of Water Chemistry
The reactions just described can be applied to calculate the equilibrium dissolved lead con-
centration as a function of water chemistry. Figure A.2 illustrates the dissolved lead concentration
as a function of pH for waters with two different levels of DIC. Figure A.3 shows the solid and
dissolved species present when orthophosphate has been added to water with low DIC.
The expected solid phase, as well as the presence of any solid phase, at equilibrium depends
upon the composition of the aqueous phase. In the absence of orthophosphate, cerussite and hydro-
cerussite will form, with hydrocerussite favored at higher pH (Figure A.2). In the presence of
both DIC and orthophosphate, hydroxylpyromorphite becomes the expected phase as DIC and/
or pH decrease (Figure A.3). At sufficiently high pH, lead hydroxide (Pb(OH)2(s)) precipitation is
predicted.
The lead(II) phases just discussed are formed at oxidation-reduction potentials of aer-
ated solutions and solutions treated with chloramines. At the higher ORP values established when
hypochlorite is present, lead(IV) oxide forms as the most stable phase. Figure A.4a–b illustrates
the favored solid and dissolved species as a function of both pH and ORP. When lead(IV) is the
dominant oxidation state as controlled by free chlorine, the dissolved lead concentrations can be
very low (Figure A.5). Changes in ORP that result from changes in the residual disinfectant or in
disinfectant demand can lead to transformations from low solubility PbO2(s) to higher solubility
lead(II) carbonate and phosphate phases (Schock and Giani 2004). Oxidation-reduction chemistry
can also explain the higher dissolved lead concentrations found when chloramines are present
instead of free chlorine (Vasquez et al. 2006). Despite the now widespread observation of lead(IV)
oxides in pipe scales, the chemical information necessary to determine solubility and reaction path-
ways is insufficiently developed. Continuing research to better determine the solubility product

a) b)
Figure A.3 Dissolved lead concentration and solid phases calculated to be present for a sys-
tem initially containing 1 g/L hydroxylpyromorphite, 10 mg/L dissolved inorganic carbon,
and (a) 1 mg/L or (b) 4 mg/L of dissolved orthophosphate
a) b)
1.6 1.6
PbO2(s) PbO 2(s)
1.2 1.2
PbO32-
PbO 32-
0.8 0.8
Pb5(PO4) 3OH(s)
Pb2+ Pb2+
0.4 0.4
PbCO3(aq)
EH (V)
EH (V)
Pb(OH)42-
Pb(OH)42-
0.0 PbHCO3+ Pb(OH) 2(s) 0.0 Pb(OH) 2(s)
-0.4 -0.4
-0.8 Pb(s) -0.8 Pb(s)
-1.2 -1.2
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
pH pH
Figure A.4 Predominance areas of lead(0)/lead(II)/lead(IV) system for 15 μgL total lead,
30 mg/L dissolved inorganic carbon, and (a) no orthophosphate or (b) 3 mg/L orthophos-
phate. Dashed lines are the stability limits of water.

-5
-7
-9
PbO2
-11
Log[Pb]diss (M)
-13
-15
-17
-19
-21
-23
-25
4 5 6 7 8 9 10 11
pH
Figure A.5 Dissolved lead concentration in equilibrium with lead(IV) oxide at sufficiently
oxidizing conditions that all lead is present as lead(IV)
and complexation equilibrium constants that determine lead(IV) oxide solubility is needed, as is
research to determine the factors that control which lead(IV) solids are formed (Schock and Giani
2004).
DISSOLUTION RATES OF CORROSION PRODUCTS
Information on the dissolution rates of lead-containing solids is necessary to complement

knowledge of the equilibrium solubility. The water present in a drinking water distribution will
often not be in equilibrium with the solid phases present in the pipe. Estimates of lead concentra-
tions in lead service lines based on equilibrium solubility are qualitatively good, but they are usu-
ally not quantitatively accurate. Equilibrium-based models tend to overpredict lead concentrations.
Such models are limited by the accuracy of available equilibrium constants, transitions between
scale types, by-product release, and reaction kinetics (Edwards et al. 1999; Vasquez et al. 2006).
For systems without equilibrium between the solid and dissolved phase, the rate of dissolution and
the contact time between the water and the lead-containing solid will determine the dissolved lead
concentration. Differences between lead concentrations during times of flowing water and during
stagnation periods can be significant. For example, lead concentrations in new pipe were low dur-
ing flowing conditions but exceeded 15 μg/L after 8 hours of stagnation, even in the presence of
inhibitors (Hozalski et al. 2005). During stagnant periods, dissolved lead concentrations in lead
pipe approach equilibrium exponentially with time. The largest increases occur within the first
24 hours. Release rate data could be fit using a radial diffusion model with a diffusion barrier term,
but this model may not be mechanistically accurate (Lytle and Schock 2000).

Theoretical Basis of Dissolution Rates
Dissolution rates of precipitated solids are functions of properties of both the dissolving
solid and the solution in which it is dissolving. A general dissolution rate law (Eq. A.12) has been
developed based on chemical thermodynamics (Lasaga 1998).
Rate = k0 {H +} nH+ {i} ni ]1 − 10 SI g

Ω
(A.12)
[Pb 2 +] 3 [CO 23 −] 2 [H +] − 2
SI = log d IAP n = log e o (A.13)
The rate is expressed in units of mol m–2 h–1 and is applicable for a given temperature. The dis-
solution rate constant k0 is also in mol m–2 h–1. The effects of pH are included by considering the
proton activity {H+} and the order of the reaction with respect to H+ (nH+). Additional dissolution
enhancing or inhibiting species are expressed as {i} with ni as their reaction orders; possible inhib-
iting or enhancing species include orthophosphate, polyphosphates, carbonate, and complexing
ligands. The last term includes the saturation index (SI), which expresses the ratio of the ion activ-
ity product of a solid to its equilibrium solubility product, shown in Equation A.13 for the case of
hydrocerussite. The SI term accounts for the extent of saturation (i.e., supersaturated vs. undersatu-
rated) of the solution with respect to the dissolving solid. The term Ω is an empirical coefficient
to account for the non-elementary nature of dissolution reactions. The greater the extent of under-
saturation (i.e., SI is more negative), the higher the dissolution rate. As the solution approaches
equilibrium with respect to the solubility of the dissolving solid, SI approaches zero and the net
dissolution rate approaches zero. For supersaturated solutions, SI is positive, and the dissolution
rate becomes negative (i.e., it becomes a precipitation rate). Note that pH, orthophosphate, and
carbonate can affect the dissolution rate by two possible mechanisms: (1) directly as inhibiting or
enhancing species and (2) by altering the solution saturation state.
Dissolution Rates of Carbonate and Phosphate Solids
Carbonate Solids With and Without Lead
Carbonate solids are among the most rapidly soluble minerals (Stumm and Morgan
1996), and their dissolution rates increase significantly with decreasing pH at acidic conditions.
Carbonate dissolution rates are often found to follow Equation A.12 in accounting for enhancing
and inhibiting species and the dependence on solution saturation. Table A.2 summarizes several
recent studies of the dissolution rates of carbonate minerals. A comparative study of different cal-
cium carbonate solids determined rate constants (k in equation A.12) of 6.6·10–3 mol m–2 h–1 and
9.7·10–3 mol m–2 h–1 for calcite and the biogenic carbonate in shells, respectively (Cubillas et al.
2005). The exponent on the term for the effect of saturation state (Ω in Equation A.12) was 1.25
for calcite and 0.86 for the biogenic carbonate in shells (Cubillas et al. 2005). Slightly faster rates
were observed for calcite in seawater with a rate constant of 4.7·10–2 mol m–2 h–1 and exponent
of 1.5 on the saturation state term (Gledhill and Morse 2004). Because of the dependence on the
saturation state and the increasing solubility of carbonate minerals with decreasing pH, the use
of Equation A.12 can predict the increase in dissolution rate that occurs with decreasing pH. The

Table A.2
Summary of selected previous studies of carbonate dissolution rates
Dissolution Rate
Solid Conditions (mol m–2 h–1) Ref.*
Calcite varied, pH freshwater 6.6·10–3·(1–10SI)1.25 1
CaCO3(s)
Biogenic Aragonite varied pH, freshwater 9.7·10–3·(1–10SI)0.86 1
CaCO3(s)
Calcite seawater 4.7·10–2·(1–10SI)1.5 2
CaCO3(s)
Calcite pH 7 ~10–3 3
CaCO3(s)
PbCO3(s) pH 5–8 ~10–4 4
*1 - (Cubillas et al. 2005); 2 - (Gledhill and Morse 2004); 3 - (Chou et al. 1989); 4 - (Pokrovsky and Schott 2002).
rate is much less dependent on pH at neutral to basic pH. A similar dependence on pH has been
determined by considering the rate constants for a set of three elementary reactions that comprise
the overall reaction and the related empirical Equation A.14 (Chou et al. 1989).
Rate = k1 {H +} nH+ + k2 {H2 CO3 *} nH2CO3 * + k3

(A.14)
For calcite, these equations are consistent with a dissolution rate on the order of 10–3 mol m–2 h–1
at pH 7 and that increases with decreasing pH. Similar trends were observed for magnesium and
barium carbonates, but the dissolution rate of the magnesium carbonate was four orders of magni-
tude lower than that of calcite.
Pokrovsky and co-workers have further developed the rate dependence on saturation
and surface speciation for metal carbonates (Pokrovsky and Schott 1999; Pokrovsky et al. 1999;
Oelkers et al. 2002). The rates fit Equation A.12 by considering the term for solution saturation
and two enhancing species, which were identified as the surface species that are most vulnerable
to dissolution. Their model can account for both dissolution and precipitation with the same rate
equation. They have developed this model in detail for magnesite (MgCO3(s)) (Pokrovsky and
Schott 1999; Pokrovsky et al. 1999; Oelkers et al. 2002) and also applied it to a series of divalent
metal carbonates (Pokrovsky and Schott 2002). In the overview of divalent metal carbonates, dis-
solution rates increased in the sequence Ni < Mg < Co < Fe < Mn < Zn < Cd <Sr < Ca ≈ Ba ≈ Pb.
Dissolution rates of PbCO3(s) were on the order of 10–4 mol m–2 h–1 (Pokrovsky and Schott 2002).
Temperature affects dissolution rates as expressed by the Arrhenius equation (Eq. A.15),
Ea
k = A $ e RT (A.15)
k ^T1 h E
ln d n = a c 1 − 1 m (A.16)
k ^T2 h R T1 T2

Table A.3
Summary of selected previous studies of phosphate dissolution rates
Dissolution Rate
Solid Conditions (mol m–2 h–1) Ref.*
Pyromorphite pH 2–7 3.9·10–3·{H+}0.65(1–10SI) 1
Pb5(PO4)3Cl(s) pH 7 and undersaturated 1.1·10–7
Fluorapatite pH 6–8.5 2.35·10–7 2
Ca5(PO4)3F(s) pH < 6 2.1·10 ·{H+}0.81(1–10SI)
–2
Syn. Hydroxylapatite pH 6–8.5 1.6-1.9·10–7 3

Ca5(PO4)3OH(s) pH 2.2–2.9 2.4-8.4·10–7
*1 - (Xie and Giammar 2007); 2 - (Guidry and Mackenzie 2003); 3 - (Valsami-Jones et al. 1998).
In which k is the rate constant, A is a pre-exponential Arrhenius term, and Ea is the activation
energy of the reaction. Reaction rates at two temperatures can be compared by rearrangement of
Equation A.15 for the two temperatures (Eq. A.16). The activation energy of calcite dissolution at
acidic pH was found to be 19 kJ mol–1 (Alkattan et al. 1998), which would lead to a 23% decrease
in the dissolution rate constant as the temperature decreases from 25°C to 15°C.
Phosphate Solids With and Without Lead
Research specific to lead phosphate solubility and dissolution-precipitation kinetics has

been investigated in previous studies that were motivated by an interest in precipitating lead phos-
phate minerals for in situ immobilization of lead in soil or groundwater (Nriagu 1973; Ma et al.
1995; Hettiarachchi et al. 2000; Yang et al. 2001; Cao et al. 2002; Ryan et al. 2004). ��
The dissolu-
tion rates of phosphate minerals have also been studied previously out of interest in medical and
dental applications since the calcium phosphate apatite (Ca5(PO4)3(F,OH,Cl)(s)) is the inorganic
component of teeth and bones.
Of the lead phosphate minerals, chloropyromorphite (Pb5(PO4)3Cl(s)), a phase closely
related to hydroxylpyromorphite, has received the most attention. Precipitation of chloropyromor-
phite occurs rapidly and is usually homogeneous (i.e., solids nucleate directly from solution without
needing a pre-existing surface) (Lower et al. 1998; Manecki et al. 2000). In contrast to precipita-
tion, the dissolution rates of lead phosphates have not received much attention. Scheckel and Ryan
observed slower dissolution of chloropyromorphite for materials that had aged for longer periods
of time or that were reacted at higher pH (Scheckel and Ryan 2002). Recent work on the dissolu-
tion rates of chloropyromorphite observed that dissolution rates increase with decreasing pH and
decreasing solution saturation. In this work they developed a model based on Equation A.12, with
k0 of 0.0039 mol m–2 h–1 and nH+ of 0.65, that could successfully predict the measured dissolution
rates. At a pH of 7 and a highly undersaturated solution, this model predicts a dissolution rate of
1.1·10–7 mol m–2 h–1 (Xie and Giammar 2007).
The dissolution rates of calcium phosphate apatite minerals have been measured as func-
tions of pH, solution saturation, and the composition of the solid phase. A summary of recent
investigations of phosphate dissolution rates is included as Table A.3. The dissolution rate of
apatite generally increases with decreasing pH in the acidic pH range, but it can be relatively inde-
pendent of pH at neutral pH. For fluorapatite dissolution, Guidry and Mackenzie observed a pH-
independent dissolution rate of 2.35·10–7 mol m–2 h–1 for pH 6-8.5, and at lower pH, dissolution

followed Equation A.12 with k0 of 0.021 mol m–2 h–1 and nH+ of 0.81 (Guidry and Mackenzie
2003). For synthetic hydroxyapatite, dissolution rates of 1.6–1.9·10–7 mol m–2 h–1 were found
at pH 6.5–6.7 and faster rates of 2.4–8.4·10–6 mol m–2 h–1 at pH 2.2–2.9 (Valsami-Jones et al.
1998). Qualitatively similar pH-dependent dissolution rates have been measured for the disso-
lution of specific crystal faces of a natural apatite (Tseng et al. 2006). The rate of apatite dis-
solution has been observed to decrease with increasing solution saturation, which is in accord
with Equation A.12 (Guidry and Mackenzie 2003), and far from saturation the dissolution rates
become almost independent of the saturation state (Tang et al. 2003b). The dissolution rate can
depend on the composition of the material. Fluorapatite dissolved more rapidly than did a carbon-
ated fluorapatite (Guidry and Mackenzie 2003) or a hydroxylapatite (Valsami-Jones et al. 1998). A
carbonated hydroxylapatite dissolved faster than a stoichiometrically pure hydroxylapatite (Tang
et al. 2003a). For compositionally similar materials, dissolution rates are lowest for more crys-
talline materials (Fulmer et al. 2002). While dissolution rates will be proportional to the surface
area of the solid, the particle size may also influence rates through other means. Recent work
observed slower dissolution rates for nanoparticles of hydroxylapatite than for larger particles of
similar compositions, which was explained by the limitation of forming initial dissolution pits on
nanoparticles (Tang et al. 2004).
The dissolution rates of phosphate solids decrease with decreasing temperature. The acti-
vation energy of fluorapatite has been measured as 34.8 kJ mol–1 at pH 3.0 (Guidry and Mackenzie
2003). When used in Equations A.15 and A.16, this value would result in a 39% decrease in the
dissolution rate constant with a 10°C drop in temperature.
Additional factors that can affect the overall dissolution rate of phosphates are the rate-
limiting step in dissolution, the stoichiometry of dissolution, and other enhancing and inhibiting
effects. Dissolution rates can be either controlled by the rate of reactions at the surface, the rate of
transport of dissolved species from the interface, or a combination of these processes; for the rela-
tively slow dissolution of phosphate minerals, the dissolution rate has been found to be surface-
controlled (Hsu et al. 1994; Guidry and Mackenzie 2003; Tang et al. 2003b). The dissolution of
apatite solids is initially incongruent with the release of calcium to solution in excess of its stoi-
chiometric ratio to phosphate, although the dissolution becomes congruent with increasing reaction
time (Schaad et al. 1997; Valsami-Jones et al. 1998; Guidry and Mackenzie 2003). The adsorption
of cations, including Pb2+ and Ca2+, can inhibit dissolution (Schaad et al. 1997). At higher load-
ings of lead, the complete exchange of lead for calcium has occurred through the dissolution of
hydroxyapatite followed by the heterogeneous nucleation and epitaxial growth of a lead phosphate
(Valsami-Jones et al. 1998). Dissolution can also be enhanced by the presence of organic ligands,
such as citric acid which accelerated hydroxylapatite dissolution (Tang et al. 2003a).
RECOMMENDATIONS FOR FUTURE RESEARCH
Critical knowledge and information that are important for evaluating the effect of dissolu-
tion-precipitation reactions in lead service lines on dissolved lead concentrations are still missing.
Topics that would benefit from additional research are (1) improved determinations of equilibrium
constants, (2) the identification of the specific phases, including metastable phases, that form in
pipe scales, and (3) the dissolution rates of lead corrosion products.
Information as fundamental as equilibrium constants remains poorly constrained for some
systems. This is particularly true for the lead(IV) oxides scrutinyite and plattnerite, which have
almost no direct measurements of their solubility. Equally relevant to the solubility of lead(IV)

solids are the equilibrium constants for dissolved lead(IV) complexes with hydroxide and possibly
with carbonate. Even for materials for which equilibrium constants are available, such as hydro-
cerussite, the available constants may differ by more than an order of magnitude. Experiments con-
ducted at conditions relevant to distribution system water chemistry can help improve estimates of
equilibrium solubility.
The identities of the actual solid phases that form in a system are often poorly understood.
Often the thermodynamically most stable phase, which would be predicted from an equilibrium
calculation, does not form first. Instead, metastable phases of higher solubility may precipitate
first and persist indefinitely. The lead phosphate system is one in which the identification of the
exact phases often remains unclear. Over a small range of conditions (e.g., pH), different lead
phosphate phases are predicted to be dominant. Further, lead phosphates can incorporate ions that
are not present in the pure end member solids, and these substitutions can affect the solubility. For
example, carbonate can substitute into the calcium phosphate structures and increase the solubil-
ity; similar substitutions may be expected for lead phosphate structures. Conversely, the presence
of chloride can promote the formation of chloropyromorphite instead of hydroxylpyromorphite,
which can reduce the equilibrium dissolved lead concentrations by several orders of magnitude.
Accurate information regarding the dissolution rates of lead corrosion products is neces-
sary to estimate the dissolved lead concentrations in tap water. Based on water chemistry and
contact time of the water with the corrosion products, dissolved lead concentrations can then be
estimated. Dissolution rates must be known for important solid phases as a function of relevant
water chemistry parameters. The present study is advancing our knowledge of rates for three
important corrosion products.

APPENDIX B
EQUILIBRIUM CONSTANTS AND REACTIONS
Equilibrium constants and reactions used in calculations in this work can be found in
Tables B.1 to B.4.
The equilibrium constant for Reaction 22 was calculated by first determining the Gibbs
free energy of the reaction by the summation of the molar Gibbs free energies of formation (G0f,i)
for each component (i), Equations B.1 and B.2. The Gibbs free energies of formation are listed in
Table B.2.
k
ΔGr = / G 0f, i Ni (B.1)
i=1
where DGr = Gibbs free energy of reaction (r), joules (J/mol)

G0f,i = Gibbs free energy of formation of species (i), joules per mole (J/mol)
Ni = stoichiometric coefficient
ΔGr, 20 = μPbO 23 − + 6 ) μH + − μPb 4 + − 3 ) μH2 O −= 131, 481 J (B.2)
The equilibrium constant can then be calculated by Equation B.3:
ΔGr
log Keq =− (B.3)
2.303RT
where R = equilibrium gas constant, joules per mole per Kelvin (J/mol·K)
T = temperature, Kelvin (K)
131, 481
log (Keq) =− −=− 23.04 (B.4)
2.303RT
The equilibrium constant for Reaction 23 was calculated by the summation of the chemical
potentials in an analogous manner to Reaction 22.
The equilibrium constant for Reaction 24 was calculated by adding Reactions 18 and 19.
The equilibrium constants for Reactions 25 and 26 were made by similar combinations of reac-
tions in Table B.1.
The Gibbs free energy of reaction for Reaction 37 was determined by the summation of
free energies of formation (Table B.2), Equation B.1. The equilibrium constant was then deter-
mined by Equation B.3.
113

Table B.1
Equilibrium constants for aqueous species
# Reaction Log K Source
1 H2O = H+ + OH– –14.00 MINEQL+
2 CO2(g) + H2O = H2CO3* –1.46 MINEQL+
3 H2CO3* = 2H+ + CO32– –16.68 MINEQL+
4 HCO3– = H+ + CO32– –10.33 MINEQL+
5 H3PO4 = 3H+ + PO43– –21.70 MINEQL+
6 H2PO4– = 2H+ + PO43– –19.56 MINEQL+
7 HPO42– = H+ + PO43– –12.35 MINEQL+
8 Pb2+ + H2O = PbOH+ + H+ –7.60 MINEQL+
9 Pb2+ + 2H2O = Pb(OH)20 + 2H+ –17.12 Benjamin
10 Pb2+ + 3H2O = Pb(OH)3– + 3H+ –28.06 Benjamin
11 Pb2+ + 4H2O = Pb(OH)42– + 4H+ –39.70 Benjamin
12 Pb2+ + CO32– = PbCO30 6.48 MINEQL+
13 Pb2+ + 2CO32– = Pb(CO3)22– 9.94 MINEQL+
14 Pb2+ + CO3– + H+ = PbHCO3+ 13.20 MINEQL+
15 2Pb2+ + 3H2O = Pb2(OH)3+ + 3H+ –6.40 MINEQL+
18 HOCl + 2e– + H+ = Cl– + H2O 50.20 Benjamin
19 Pb4+ + 2e– = Pb2+ 28.64 Benjamin
20 O2(aq) + 4H+ +4e– = 2H2O 86.00 Benjamin
21 2H+ +2e– = H2(aq) 3.10 Benjamin
22 Pb4+ + 3H2O = PbO32– + 6H+ –23.04 Calculated
23 Pb4+ + 4H2O = PbO44– + 8H+ –63.80 Calculated
Benjamin = (Benjamin 2002)
MINEQL+ = (Schecher and Mcavoy 1998)

Appendix B: Equilibrium Constants and Reactions | 115
Table B.2
Chemical potentials for various aqueous species
Species G0f,i (J/mol) Source
Pb3O4(s) –601,200 Benjamin
Pb2O3(s) –411,769 Pourbaix
PbO2(s) –218,987 Pourbaix
Pb2+ –24,309 Pourbaix
HPbO2– –338,898 Pourbaix
Pb4+ 302,498 Pourbaix
PbO32– –277,562 Pourbaix
PbO44– –282,084 Pourbaix
H+ 0.00 Benjamin
OH– –157,300 Benjamin
H2O –237,180 Benjamin
Pourbaix = (Pourbaix 1974)
Table B.3
Equilibrium constants for select reactions of aqueous species
# Reaction Log K Reactions
24 HOCl + Pb2+ + H+ = Cl– + Pb4+ + H2O 21.56 15 + 16
25 HOCl + Pb2+ + 2H2O = PbO32– + Cl– + 5H+ –1.48 15 + 19
26 HOCl + Pb2+ + 3H2O = PbO44– + Cl– + 7H+ –42.24 15 + 20

Table B.4
Solubility product of select lead solids
# Solid Reaction Log K Source
28 Massicot PbO(s) + H2O = Pb2+ + 2OH– –15.09 Benjamin
29 Litharge PbO(s) + H2O = Pb2+ + 2OH– –15.30 MINEQL+
30 Pb(OH)2(s) Pb(OH)2(s) + 2H+ = Pb2+ + 2H2O 8.15 MINEQL+
31 Cerussite PbCO3(s) = Pb2+ + CO32– –13.13 Benjamin
32 Hydrocerussite Pb3(CO3)2(OH)2(s) + 2H+ = 3Pb2+ + 2CO32– + 2H2O –18.77 MINEQL+
33 Pb3(PO4)2(s) Pb3(PO4)2(s) = 3Pb2+ + 2PO43– –44.50 Benjamin
34 PbHPO4(s) PbHPO4(s) = Pb2+ + PO43– + H+ –37.80 MINEQL+
35 Hydroxyl- Pb5(PO4)3OH(s) + H+ = 5Pb2+ + 3PO43– + H2O –62.79 MINEQL+
pyromorphite
36 Plattnerite Pb(IV)O2(s) + 4H+ = Pb4+ + 2H2O 20.96 MINTEQA2
37 Scrutinyite Pb(IV)O2(s) + 4H+ = Pb4+ + 2H2O –8.26 Calculated
MINTEQA2 = (Allison 2003)

APPENDIX C
RELATIONSHIP AMONG ALKALINITY, pH, AND DISSOLVED
INORGANIC CARBON
Both the carbonate ion concentration and alkalinity are strongly pH-dependent. The car-
bonate ion concentration can be determined from the pH and dissolved inorganic carbon concen-
tration (DIC) (C.1).
DIC = [H2CO3*] + [HCO3–] + [CO32–] (C.1)
nate the alkalinity (C.2), then DIC can be calculated from pH and alkalinity.
Carbonate Alkalinity = –[H+] + [OH–] + [HCO3–] + 2[CO32–] (C.2)
Table C.1 summarizes the alkalinity for the combinations of pH and DIC studied in this
investigation.
Table C.1
Alkalinity for combinations of pH and DIC investigated
Alkalinity in mg/L as CaCO3
DIC
(mg C/L) pH 7.5 pH 8.5 pH 10
0 0.0 0.2 5.0
10 39.0 42.1 60.0
50 194.9 210.1 279.7
117

APPENDIX D
TRACER STUDY OF CONTINUOUS-FLOW STIRRED REACTOR
The tracer study was performed by introduction of a 30 μg/L dissolved lead solution into a
stirred reactor containing ultrapure water at time zero. A flow rate of 2.83 mL/min was set to pro-
vide a hydraulic residence time of 30 minutes. After two hours the influent solution was switched
from the 30 μg/L lead solution to ultrapure water. Ideal well-mixed flow-through reactor behavior
for both the introduction of the lead solution and the change to ultrapure water was modeled using
the Equation D.1 and the results are shown in Figure D.1.
t
C (t) = Cin (1 − e tres ) (D.1)
40
30
Dissolved Lead (ppb)
20
10
0
0 50 100 150 200 250 300
Time (min)
Figure D.1 Dissolved lead concentration in a well-mixed flow-through reactor with a 30 min-
ute hydraulic residence time. Initial [Pb]diss in the reactor is 0 μg/L. Influent [Pb]diss at time 0
is 30 μg/L. Influent changed after 2 hours to a water solution where [Pb]diss is 0 μg/L. Mod-
eled well-mixed flow-through reactor behavior shown as solid lines.
119

APPENDIX E
CALCULATING EQUILIBRIUM LEAD CONCENTRATIONS FOR
HYDROCERUSSITE EXPERIMENTS
The dissolved lead concentration in equilibrium with hydrocerussite is needed to calcu-

late the dissolution rate of hydrocerussite. The calculation is complicated by the contribution of
carbonate from the dissolution of hydrocerussite and the initial dissolved inorganic carbon in the
influent solution that comes from the dissolution of carbon dioxide in the air. In two different dis-
solution reactions the influent reservoirs were open with respect to exchange with the atmosphere
(pH = 7.5, DIC = 2.19 mg/L and pH = 8.5, DIC = 20.9 mg/L), but the flow-through reactors them-
selves were closed system with respect to gas exchange. For every other dissolution experiment,
the influent was prepared as a closed system. Equilibrium constants and reactions used in calcula-
tions can be found in Appendix B and are included here in condensed form in Table E.1. The set of
equilibrium constants used can affect the calculated equilibrium concentration by about an order
of magnitude. The calculations are most significantly affected by the choice of the solubility prod-
uct for hydrocerussite. The total dissolved inorganic carbon in the system is a sum of the carbon
contributed initially in the influent and that from the dissolution of hydrocerussite (Equation E.1).
HC
T + CT
CT = C int (E.1)
where CT = total dissolved inorganic carbon concentration, moles per liter (mol/L)
CTint =dissolved inorganic carbon concentration initially contributed by the influent,
moles per liter (mol/L)
CTHC =dissolved inorganic carbon concentration contributed by the dissolution of
hydrocerussite, moles per liter (mol/L)
If the system is open to the atmosphere, then the initial concentration in the influent can be calcu-
lated as a function of pH by Equation E.2.
Air K3 $ K2 $ PCO2 K4 $ K3 $ K2 $ PCO2

C int
T = C T = K2 $ PCO2 + + (E.2)
[H +] [H +] 2
where [H+] = concentration of aqueous species H+, moles per liter (mol/L)
Kn = equilibrium constant of reaction n (Table E.1)
PCO2 = partial pressure of carbon dioxide in the air, bar (bar)
The carbonate concentration from the dissolution of hydrocerussite (Pb3(CO3)2OH(s)) is a fraction

of the total amount of lead (Pb) dissolved (Equation E.3).
C THC = 2 [Pb] diss (E.3)

3
where [Pb]diss = concentration of total dissolved lead species, moles per liter (mol/L)
121

Table E.1
Equilibrium constants for aqueous species
# Reaction Log K Source
1 H2O → H+ + OH– –13.83 Benjamin
2 CO2(g) + H2O → H2CO3* –1.459 MINEQL+
3 H2CO3* → 2H+ + CO32– –16.68 MINEQL+
4 HCO3– → H+ + CO32– –10.33 MINEQL+
5 Pb2+ + H2O → PbOH+ + H+ –7.597 MINEQL+
6 Pb2+ + 2H2O → Pb(OH)20 + 2H+ –17.12 Benjamin
7 Pb2+ + 3H2O → Pb(OH)3– + 3H+ –28.06 Benjamin
8 Pb2+ + 4H2O → Pb(OH)42– + 4H+ –39.70 Benjamin
9 Pb2+ + CO32– → PbCO30 6.478 MINEQL+
10 Pb2+ + 2CO32– → Pb(CO3)22– 9.38 MINEQL+
11 Pb2+ + CO3– + H+ → PbHCO3+ 13.20 MINEQL+
12 2Pb2+ + 3H2O → Pb2(OH)3+ + 3H+ –6.397 MINEQL+
15 Pb3(CO3)2(OH)2(s) + 2H+ → 3Pb2+ + 2CO32– + 2H2O –18.77 MINEQL+
The total dissolved lead is the sum of all dissolved lead species, which includes hydroxide
complexes and complexes with carbonate in addition to Pb2+ (Equations E.4–8).
[Pb] diss = [Pb 2 +] + Pb (hydroxides) + Pb (carbonates) (E.4)
Pb (hydroxides) = [PbOH +] + [Pb (OH) 02] + [Pb (OH) −3 ] + [Pb (OH) 42 −] (E.5)
K5 K6 K7 K8
Pb (hydroxides) = [Pb 2 +] $ e + + + 2 + + 3 + o (E.6)
[H ] [H ] [H ] [H +] 4
Pb (carbonates) = [PbCO 03] + [Pb (CO3) 2 −] + [PbHCO +3 ] (E.7)
Pb (carbonates) = [Pb 2 +] $ _ K9 [CO 23 −] + K10 [CO 32 −] 2 + K11 [H +] [CO 23 −] i (E.8)
where [i] = concentration of aqueous species i, moles per liter (mol/L)

Appendix E: Calculating Equilibrium Lead Concentrations for Hydrocerussite Experiments | 123
5.0E-06
4.5E-06 pH = 7.5
pH = 8.5
4.0E-06 pH = 10.0
3.5E-06
3.0E-06
Pbdiss (M)
2.5E-06
2.0E-06
1.5E-06
1.0E-06
5.0E-07
0.0E+00
0 5 10 15 20 25 30 35 40 45 50
DIC (mg/L)
Figure E.1 The effect of influent dissolved inorganic carbon on the total dissolved lead con-
centration (Pbdiss) in equilibrium with hydrocerussite. ● pH = 7.5, DIC = 2.19 mg/L. ○ pH =
8.5, DIC = 20.9 mg/L.
The solubility product of hydrocerussite is used to determine the lead ion (Pb2+) concentration
(Equation E.9).
6Pb 2 +@3 $ 6CO 23 −@2

Ksp = K15 = (E.9)
6H +@2
The equilibrium concentration of lead can be calculated by simultaneously solving
Equations E.1–E.9 using an iterative process. A routine for executing this iterative solution has
been developed using the Solver feature in Microsoft Excel. The calculations can also be made
using MINEQL+, which does allow for the modification of equilibrium constants from those in
its default database. When using MINEQL+, the total (including solids) lead and inorganic carbon
concentrations are set to fixed values determined from the concentration of solids loaded into the
reactor and the initial dissolved inorganic carbon contributed by the atmosphere.
The effect of influent dissolved inorganic carbon (DIC) on the amount of total dissolved
lead is shown in Figure E.1. As the dissolved inorganic carbon concentration increases the equilib-
rium concentration [Pb]diss of total dissolved lead increases. The increase in dissolved lead concen-
tration is directly related to the increase in concentrations of lead carbonates (PbCO3, Pb(CO3)22–,
and PbHCO3). As the pH of influent solution increases the total amount of dissolved inorganic
carbon increases. In the case of the two open system influents (pH = 7.5, DIC = 2.19 mg/L and pH
= 8.5, DIC = 20.9 mg/L) the [Pb]diss (or equilibrium concentration, Ceq) for the lower pH is slightly
greater than the higher pH (8.51·10–7 M and 1.39·10–6).

APPENDIX F
ORDER OF CONDUCTING EXPERIMENTS WITH LEAD PIPES
Lead release experiments from scales in pipe were conducted in duplicate for conditions
to examine the effects of stagnation time, solution chemistry, and water flow on release rates. Six
pipes were selected for these experiments (Table F.1). Because the experiments themselves could
potentially alter the pipe scales, the order in which conditions were studied was not the same for
each of the pipes (Table F.2) so that any effects of the order of conducting experiments could be
separated from actual effects of the experimental variables.
Table F.1
Schedule of experiments with laboratory-developed scales in new lead pipes
Solution 1 Solution 2 Solution 3
Date Pipe 8 Pipe 9 Pipe 6 Pipe 7 Pipe 11 Pipe 12
Sept. 22 Conditioning Conditioning Conditioning Conditioning Conditioning Conditioning
Oct. 1 Conditioning Conditioning Conditioning Conditioning Conditioning Conditioning
Oct. 14 0.1 m/s, s.t. 0.1 m/s, s.t. 0.1 m/s, s.t.
1 hr, s.t. 4 hr, 1 hr, s.t. 4 hr, 1 hr, s.t. 4 hr,
s.t. 8 hr s.t. 8 hr s.t. 8 hr
Oct. 16 velocity 0 m/s velocity 0 m/s velocity 0 m/s velocity 0 m/s velocity 0 m/s velocity 0 m/s
Oct. 21 0.1 m/s, s.t. 24 0.1 m/s, s.t. 0.1 m/s, s.t.
hr, 48 hr 24 hr, 48 hr 24 hr, 48 hr
Oct. 23 0.1 m/s, s.t. 0.1 m/s, s.t. 0.1 m/s, s.t.
24 hr, 48 hr 24 hr, 48 hr 24 hr, 48 hr
Oct. 28 velocity 0 m/s velocity 0 m/s velocity 0 m/s velocity 0 m/s velocity 0 m/s velocity 0 m/s
Oct. 30 0.1 m/s, s.t. 0.1 m/s, s.t. 0.1 m/s, s.t. 1 hr,
1 hr, s.t. 4 hr, 1 hr, s.t. 4 hr, s.t. 4 hr, s.t. 8 hr
s.t. 8 hr s.t. 8 hr
125

Table F.2
Order of experiments with pipe sections from an actual lead service line
Pipe sections
Water quality 2 Water quality 1 Water quality 3
Conditions 1 4 2 5 3 6
No Flow X X X
Low Flow X X X X X X
No Flow X X X
High Flow X X X X X X

APPENDIX G
RESULTS OF EXPERIMENTS WITH LEAD SERVICE LINES
The conditions and results of lead release experiments are summarized in Table G.1.
Table G.1
Conditions and results of lead release experiments
Influent composition Flow Reaction Total Diss.
Pipe DIC Orthophosphate Measured velocity time lead lead
number pH (mg/L as C) (mg/L as P) pH (m/s) (hour)* (nM) (nM)†
1 7.5 10 1 7.50 0.00 0 2 1
1 7.5 10 1 7.77 0.00 1 577 256
1 7.5 10 1 7.85 0.00 4 1533 692
1 7.5 10 1 7.86 0.00 8 1453 743
1 7.5 10 1 7.69 0.00 24 978 681
1 7.5 10 1 7.78 0.00 48 1050 801
1 7.5 10 1 7.48 0.05 0 0 0
1 7.5 10 1 7.95 0.05 2 262 241
1 7.5 10 1 7.86 0.05 3 221 213
1 7.5 10 1 7.64 0.05 6 263 213
1 7.5 10 1 7.72 0.05 10 262 290
1 7.5 10 1 7.72 0.05 26 288 318
1 7.5 10 1 7.69 0.05 50 338 333
1 7.5 10 1 7.54 0.30 0 0 0
1 7.5 10 1 8.09 0.30 2 379 159
2 8.5 50 0 8.49 0.00 0 10 3

2 8.5 50 0 8.62 0.00 1 867 982
2 8.5 50 0 8.93 0.00 4 1360 1127
2 8.5 50 0 9.15 0.00 8 1381 1321
2 8.5 50 0 8.98 0.00 24 1029 1165
2 8.5 50 0 9.14 0.00 48 1214 1056
2 8.5 50 0 8.55 0.05 0 0 0
2 8.5 50 0 8.58 0.05 2 2556 1402
2 8.5 50 0 8.58 0.05 3 2478 1876
2 8.5 50 0 8.58 0.05 6 1801 1496
2 8.5 50 0 8.59 0.05 10 1522 1296
2 8.5 50 0 8.6 0.05 26 1662 1267
2 8.5 50 0 8.62 0.05 50 1307 1213
2 8.5 50 0 8.52 0.30 0 3 0
2 8.5 50 0 8.56 0.30 2 1597 1383
(continued)
127

Table G.1 (Continued)

Influent Composition
Flow Reaction Total Diss.
Number pH (mg/L as C) (mg/L as P) pH (m/s) (hour)* (nM) (nM)†
3 10 10 0 10.02 0.00 0 5 4
3 10 10 0 9.95 0.00 1 263 380
3 10 10 0 9.52 0.00 4 524 432
3 10 10 0 9.28 0.00 8 664 537
3 10 10 0 8.82 0.00 24 483 467
3 10 10 0 8.88 0.00 48 676 613
3 10 10 0 10.01 0.05 0 0 0
3 10 10 0 9.78 0.05 2 1327 956
3 10 10 0 9.79 0.05 3 1342 1171
3 10 10 0 9.66 0.05 6 1176 838
3 10 10 0 9.54 0.05 10 1066 937
3 10 10 0 9.18 0.05 26 728 738
3 10 10 0 9.02 0.05 50 797 756
3 10 10 0 10.01 0.30 0 2 1
3 10 10 0 9.78 0.30 2 1205 1069
4 7.5 10 1 7.50 0.00 0 5 4

4 7.5 10 1 7.80 0.00 1 175 128
4 7.5 10 1 7.89 0.00 4 857 413
4 7.5 10 1 8.04 0.00 8 1041 548
4 7.5 10 1 7.66 0.00 24 2304 487
4 7.5 10 1 7.80 0.00 48 987 583
4 7.5 10 1 7.42 0.05 0 0 0
4 7.5 10 1 7.84 0.05 2 287 177
4 7.5 10 1 7.88 0.05 3 227 169
4 7.5 10 1 7.93 0.05 6 367 238
4 7.5 10 1 7.97 0.05 10 339 366
4 7.5 10 1 7.87 0.05 26 374 298
4 7.5 10 1 7.82 0.05 50 403 288
4 7.5 10 1 7.48 0.30 0 0 0
4 7.5 10 1 7.99 0.30 2 1941 1846
(continued)

Appendix G: Results of Experiments With Lead Service Lines | 129
Table G.1 (Continued)

Influent Composition
Flow Reaction Total Diss.
Number pH (mg/L as C) (mg/L as P) pH (m/s) (hour)* (nM) (nM)†
5 8.5 50 0 8.49 0.00 0 5 3
5 8.5 50 0 8.61 0.00 1 797 891
5 8.5 50 0 9.10 0.00 4 1258 1188
5 8.5 50 0 9.15 0.00 8 1709 1421
5 8.5 50 0 9.21 0.00 24 1447 1363
5 8.5 50 0 9.18 0.00 48 1280 1351
5 8.5 50 0 8.52 0.05 0 0 0
5 8.5 50 0 8.57 0.05 2 697 611
5 8.5 50 0 8.58 0.05 3 621 582
5 8.5 50 0 8.56 0.05 6 757 699
5 8.5 50 0 8.55 0.05 10 696 580
5 8.5 50 0 8.57 0.05 26 873 637
5 8.5 50 0 8.62 0.05 50 1083 835
5 8.5 50 0 8.46 0.30 0 56 55
5 8.5 50 0 8.75 0.30 2 5000 1633
6 10 10 0 10.02 0.00 0 43 0
6 10 10 0 9.69 0.00 1 254 225
6 10 10 0 9.60 0.00 4 394 404
6 10 10 0 9.55 0.00 8 526 591
6 10 10 0 9.08 0.00 24 711 659
6 10 10 0 8.86 0.00 48 1032 931
6 10 10 0 10.01 0.05 0 0 0
6 10 10 0 9.82 0.05 2 1670 862
6 10 10 0 9.8 0.05 3 1434 761
6 10 10 0 9.6 0.05 6 1321 702
6 10 10 0 9.55 0.05 10 1061 796
6 10 10 0 9.17 0.05 26 752 699
6 10 10 0 8.99 0.05 50 890 736
6 10 10 0 10.08 0.30 0 28 28
6 10 10 0 9.66 0.30 2 894 714
*For the high and low flow experiments the recirculation period starts at time = 0 and lasts for 2 hours. Stagnation
times are then added to this time. For example: 4 hours of stagnation time occur after 2 hours of recirculation for
a total of 6 hours. For the no flow experiments there was no recirculation and the reaction times reflect stagnation
times only.
†Dissolved lead concentrations were determined by filtering solutions with a 0.45 μm filter and then analyzing the
aqueous concentrations by inductively coupled plasma mass spectroscopy

REFERENCES
Alkattan, M., Oelkers, E.H., Dandurand, J.L., and Schott, J. 1998. An experimental study of calcite
and limestone dissolution rates as a function of pH from –1 to 3 and temperature from 25
to 80 degrees C. Chemical Geology, 151(1-4):199.
Allison, J.D. 2003. MINTEQA2 for Windows® Equilibrium Speciation Model, 3 Version 1.50.
American Society of Civil Engineers. 2005. Report card for America’s infrastructure.
AWWA (American Water Works Association). 2005. Managing change and unintended conse-
quences: Lead and Copper Rule corrosion control treatment. Denver, Colo.: AWWA.
Benjamin, M.M. 2002. Water Chemistry. Boston: McGraw-Hill.
Boyd, G.R., Dewis, K.M., Korshin, G.V., Reiber, S.H., Schock, M.R., Sandvig, A.M., and Giani,
R. 2008. Effects of changing disinfectants on lead and copper release. Jour. AWWA,
100(11):75.
Cantor, A.F. 2006. Diagnosing corrosion problems through differentiation of metal fractions. Jour.
AWWA, 98(1):117.
Cantor, A.F., Park, J.K., and Vaiyavatjamai, P. 2003. Effect of chlorine on corrosion in drinking
water systems. Jour. AWWA, 95(5):112.
Cao, X., Ma, L.Q., Chen, M., Singh, S.P., and Harris, W.G. 2002. Impacts of phosphate amendments
on lead biogeochemistry at a contaminated site. Environmental Science and Technology,
36(24):5296.
Chou, L., Garrels, R.M., and Wollast, R. 1989. Comparative-study of the kinetics and mechanisms
of dissolution ofcarbonate minerals. Chemical Geology, 78(3-4):269.
Churchill, D.M., Mavinic, D.S., Neden, D.G., and MacQuarrie, D.M. 2000. The effect of zinc
orthophosphate and pH-alkalinity adjustment on metal levels leached into drinking water.
Canadian Journal of Civil Engineering, 27(1):33.
Clesceri, L.S., Greenberg, A.E., and Eaton, A.D. (eds.) 1999. Standard Methods for the Examination
of Water and Wastewater. Washington, D.C.: APHA.
Cubillas, P., Kohler, S., Prieto, M., Chairat, C., and Oelkers, E.H. 2005. Experimental determination
of the dissolution rates of calcite, aragonite, and bivalves. Chemical Geology, 216(1-2):59.
Davidson, C.M., Peters, N.J., Britton, A., Brady, L., Gardiner, P.H.E., and Lewis, B.D. 2004.
Surface analysis and depth profiling of corrosion products formed in lead pipes used to
supply low alkalinity drinking water. Water Science and Technology, 49(2):49.
Dodrill, D.M., and Edwards, M. 1995. Corrosion control on the basis of utility experience. Jour.
AWWA, 87(7):74.
Dryer, D.J., and Korshin, G.V. 2007. Investigation of the reduction of lead dioxide by natural
organic matter. Environmental Science and Technology, 41(15):5510.
Dudi, A., Schock, M., Murray, N., and Edwards, M. 2005. Lead leaching from inline brass devices:
A critical evaluation of the existing standard. Jour. AWWA, 97(8):66.
Edwards, M. 2004. Controlling corrosion in drinking water distribution systems: A grand chal-
lenge for the 21st century. Water Science and Technology, 49(2):1.
Edwards, M., and Dudi, A. 2004. Role of chlorine and chloramine in corrosion of lead-bearing
plumbing materials. Jour. AWWA, 96(10):69.
Edwards, M., and McNeill, L.S. 2002. Effect of phosphate inhibitors on lead release from pipes.
Jour. AWWA, 94(1):79.
131

Edwards, M., Jacobs, S., and Dodrill, D. 1999. Desktop guidance for mitigating Pb and Cu corro-
sion by-products. Jour. AWWA, 91(5):66.
El-Egamy, S.S. 1995. Corrosion and passivation behavior of lead in aqueous carbonate media.
Bulletin of Electrochemistry, 11(9):418.
Ferris, F.G., Phoenix, V., Fujita, Y., and Smith, R.W. 2004. Kinetics of calcite precipitation
induced by ureolytic bacteria at 10 to 20 degrees C in artificial groundwater. Geochimica
et Cosmochimica Acta, 68(8):1701.
Frenkel, A.I., and Korshin, G.V. 1999. EXAFS studies of the chemical state of lead and copper in
corrosion products formed on the brass surface in potable water. Journal of Synchrotron
Radiation, 6:653.
Fulmer, M.T., Ison, I.C., Hankermayer, C.R., Constantz, B.R., and Ross, J. 2002. Measurements of
the solubilities and dissolution rates of several hydroxyapatites. Biomaterials, 23(3):751.
Giammar, D.E. 2001. Geochemistry of uranium at mineral-water interfaces: Rates of sorption-
desorption and dissolution-precipitation reactions. Environmental Engineering Science.
Giammar, D.E., and Hering, J. 2000. Dissolution and transformation rates of uranyl mineral phases.
Abstracts of Papers of the American Chemical Society, 220:153.
Giammar, D.E., and Hering, J.G. 2002. Equilibrium and kinetic aspects of soddyite dissolution and
secondary phase precipitation in aqueous suspension. Geochimica et Cosmochimica Acta,
66(18):3235.
Gledhill, D.K., and Morse, J.W. 2004. Dissolution kinetics of calcite in NaCl-CaCl2-MgCl2 brines
at 25 degrees C and 1 bar pCO(2). Aquatic Geochemistry, 10(1-2):171.
Godelitsas, A., Astilleros, J.M., Hallam, K., Harissopoulos, S., and Putnis, A. 2003. Interaction of
calcium carbonates with lead in aqueous solutions. Environmental Science and Technology,
37(15):3351.
Guidry, M.W., and Mackenzie, F.T. 2003. Experimental study of igneous and sedimentary apa-
tite dissolution: Control of pH, distance from equilibrium, and temperature on dissolution
rates. Geochimica et Cosmochimica Acta, 67(16):2949.
Hettiarachchi, G.M., Pierzynski, G.M., and Ransom, M.D. 2000. In situ stabilization of soil
lead using phosphorus and manganese oxide. Environmental Science and Technology,
34(21):4614.
Hozalski, R.M., Esbri-Amador, E., and Chen, C.F. 2005. Comparison of stannous chloride and
phosphate for lead corrosion control. Jour. AWWA, 97(3):89.
Hsu, J., Fox, J.L., Powell, G.L., Otsuka, M., Higuchi, W.I., Yu, D., Wong, J., and Legeros, R.Z.
1994. Quantitative relationship between carbonated apatite metastable equilibrium solubil-
ity and dissolution kinetics. Journal of Colloid and Interface Science, 168(2):356.
Korshin, G.V., Ferguson, J.F., and Lancaster, A.N. 2000. Influence of natural organic matter on the
corrosion of leaded brass in potable water. Corrosion Science, 42(1):53.
Korshin, G.V., Ferguson, J.F., and Lancaster, A.N. 2005. Influence of natural organic matter on the
morphology of corroding lead surfaces and behavior of lead-containing particles. Water
Research, 39(5):811.
Lasaga, A.C. 1995. Fundamental approaches in describing mineral dissolution and precipitation
rates. Chemical Weathering Rates of Silicate Minerals, 31:23.
Lasaga, A.C. 1998. Kinetic Theory in the Earth Sciences. Princeton, N.J.: Princeton University
Press.
Lin, Y.P., and Valentine, R.L. 2008a. The release of lead from the reduction of lead oxide (PbO2)
by natural organic matter. Environmental Science and Technology, 42(3):760.

References | 133
Lin, Y.P., and Valentine, R.L. 2008b. Release of Pb(II) from monochloramine-mediated reduction
of lead oxide (PbO2). Environmental Science and Technology, 42(24):9137.
Lin, Y.P., and Valentine, R.L. 2009. Reduction of lead oxide (PbO2) and release of Pb(II) in mix-
tures of natural organic matter, free chlorine and monochloramine. Environmental Science
and Technology, 43(10):3872.
Lin, Y.P., Washburn, M., and Valentine, R.L. 2008. Reduction of lead oxide (PbO2) by iodide and
formation of iodoform in PbO2/I–/NOM system. Environmental Science and Technology,
42(8):2919.
Liu, H., Korshin, G.V., Ferguson, J.F., and Jiang, W. 2006. Key parameters and kinetics of oxida-
tion of lead(II) solid phases by chlorine in drinking water. Water Practice and Technology,
1:4.
Liu, H.Z., Korshin, G.V., and Ferguson, J.F. 2008. Investigation of the kinetics and mechanisms
of the oxidation of cerussite and hydrocerussite by chlorine. Environmental Science and
Technology, 42(9):3241.
Liu, H.Z., Korshin, G.V., and Ferguson, J.F. 2009. Interactions of Pb(II)/Pb(IV) solid phases
with chlorine and their effects on lead release. Environmental Science and Technology,
43(9):3278.
Lower, S.K., Maurice, P.A., and Traina, S.J. 1998. Simultaneous dissolution of hydroxylapatite
and precipitation of hydroxypyromorphite: Direct evidence of homogeneous nucleation.
Geochimica et Cosmochimica Acta, 62(10):1773.
Lytle, D.A., and Schock, M.R. 2000. Impact of stagnation time on metal dissolution from plumb-
ing materials in drinking water. Journal of Water Supply Research and Technology-AQUA,
49(5):243.
Lytle, D.A., and Schock, M.R. 2005. Formation of Pb(IV) oxides a in chlorinated water. Jour.
AWWA, 97(11):102.
Lytle, D.A., Schock, M.R., Clement, J.A., and Spencer, C.M. 1998. Using aeration for corrosion
control. Jour. AWWA, 90(3):74.
Ma, Q.Y., Logan, T.J., and Traina, S.J. 1995. Lead immobilization from aqueous-solutions and con-
taminated soils using phosphate rocks. Environmental Science and Technology, 29(4):1118.
Manecki, M., Maurice, P.A., and Traina, S.J. 2000. Uptake of aqueous Pb by Cl–, F–, and OH–
apatites: Mineralogic evidence for nucleation mechanisms. American Mineralogist,
85(7-8):932.
Martell, A.E., and Smith, R.M. 1974. Critical Stability Constants. New York: Plenum Press.
McNeill, L.S., and Edwards, M. 2002. Phosphate inhibitor use at US utilities. Jour. AWWA,
94(7):57.
MWH. 2005. Water Treatment Principles and Design. Hoboken, N.J.: John Wiley and Sons.
Nriagu, J.O. 1973. Lead orthophosphates–II. Stability of chloropyromorphite at 25°C. Geochimica
et Cosmochimica Acta, 37:367.
Oelkers, E.H., Pokrovsky, O.S., and Schott, J. 2002. An experimental study of magnesite dissolu-
tion and precipitation rates. Geological Society of America Annual Meeting, 34.
Pokrovsky, O.S., and Schott, J. 1999. Processes at the magnesium-bearing carbonates solution inter-
face. II. Kinetics and mechanism of magnesite dissolution. Geochimica et Cosmochimica
Acta, 63(6):881.
Pokrovsky, O.S., and Schott, J. 2002. Surface chemistry and dissolution kinetics of divalent metal
carbonates. Environmental Science and Technology, 36(3):426.

Pokrovsky, O.S., Schott, J., and Thomas, F. 1999. Processes at the magnesium-bearing car-
bonates solution interface. I. A surface speciation model for magnesite. Geochimica et
Cosmochimica Acta, 63(6):863.
Pourbaix, M. 1974. Atlas of Electrochemical Equilibrium in Aqueous Solutions. Houston, Texas:
National Association of Corrosion Engineers.
Rajasekharan, V.V., Clark, B.N., Boonsalee, S., and Switzer, J.A. 2007. Electrochemistry of free
chlorine and monochloramine and its relevance to the presence of Pb in drinking water.
Environmental Science and Technology, 41:4252.
Renner, R. 2004. Plumbing the depths of D.C.’s drinking water crisis. Environmental Science and
Technology, 38(12):224A.
Rouff, A.A., Elzinga, E.J., Reeder, R.J., and Fisher, N.S. 2004. X-ray absorption spectroscopic
evidence for the formation of Pb(II) inner-sphere adsorption complexes and precipitates at
the calcite-water interface. Environmental Science and Technology, 38(6):1700.
Ryan, J.A., Scheckel, K.G., Berti, W.R., Brown, S.L., Casteel, S.W., Chaney, R.L., Hallfrisch, J.,
Doolan, M., Grevatt, P., Maddaloni, M., and Mosby, D. 2004. Reducing children’s risk
from lead in soil. Environmental Science and Technology, 38(1):18A.
Schaad, P., Poumier, F., Voegel, J.C., and Gramain, P. 1997. Analysis of calcium hydroxyapatite
dissolution in non-stoichiometric solutions. Colloids and Surfaces A–Physicochemical and
Engineering Aspects, 121(2-3):217.
Schecher, W.D., and Mcavoy, D.C. 1998. MINEQL: A chemical equilibrium modeling system.
Version 4.5. Howell, ME: Environmental Research Software.
Scheckel, K.G., and Ryan, J.A. 2002. Effects of aging and pH on dissolution kinetics and stability
of chloropyromorphite. Environmental Science and Technology, 36(10):2198.
Schock, M.R. 1989. Understanding corrosion control strategies for lead. Jour. AWWA, 81(7):88.
Schock, M.R. 1999. Internal corrosion and deposition control. In Water Quality and Treatment.
Denver, Colo.: AWWA.
Schock, M., and Giani, R. 2004. Oxidant/Disinfectant Chemistry and Impacts on Lead Corrosion.
In Proc. of the AWWA Water Quality Technology Conference.
Schock, M.R., Holldber, J., Lovejoy, T.R., and Lowry, J. 2002. California’s first aeration plants for
corrosion control. Jour. AWWA, 94(3):88.
Schock, M.R., Lytle, D.A., Sandvig, A.M., Clement, J., and Harmon, S.M. 2005a. Replacing poly-
phosphate with silicate to solve lead, copper, and source water iron problems. Jour. AWWA,
97(11):84.
Schock, M.R., Scheckel, K.G., DeSantis, M., and Gerke, T. 2005b. Mode of occurrence, treat-
ment, and monitoring significance of tetravalent lead. In Proc. of the AWWA Water Quality
Technology Conference.
Smith, B.C. 1996. Fundamentals of Fourier Transform Infrared Spectroscopy. Boca Raton, Fla:
CRC Press.
Stumm, W., and Morgan, J.J. 1996. Aquatic Chemistry. New York: John Wiley and Sons.
Switzer, J.A., Rajasekharan, V.V., Boonsalee, S., Kulp, E.A., and Bohannan, E.W. 2006. Evidence
that monochloramine disinfectant could lead to elevated Pb levels in drinking water.
Environmental Science and Technology, 40(10):3384.
Tang, R.K., Hass, M., Wu, W.J., Gulde, S., and Nancollas, G.H. 2003a. Constant composition
dissolution of mixed phases II. Selective dissolution of calcium phosphates. Journal of
Colloid and Interface Science, 260(2):379.

References | 135
Tang, R.K., Henneman, Z.J., and Nancollas, G.H. 2003b. Constant composition kinetics study of
carbonated apatite dissolution. Journal of Crystal Growth, 249(3-4):614.
Tang, R.K., Orme, C.A., and Nancollas, G.H. 2004. Dissolution of crystallites: Surface energetic
control and size effects. Chemphyschem, 5(5):688.
Tseng, W.J., Lin, C.C., Shen, P.W., and Shen, P. 2006. Directional/acidic dissolution kinetics of
(OH,F,Cl)-bearing apatite. Journal of Biomedical Materials Research Part A, 76A(4):753.
USEPA (U.S. Environmental Protection Agency). 1991. Maximum contaminant level goals
and national primary drinking water regulations for lead and copper. Final rule. Federal
Register, 56:26460.
USEPA (U.S. Environmental Protection Agency). 2007. Summary of research relevant to the D.C.
lead issue. Washington, D.C.: IOCCT Review.
Valsami-Jones, E., Ragnarsdottir, K.V., Putnis, A., Bosbach, D., Kemp, A.J., and Cressey, G. 1998.
The dissolution of apatite in the presence of aqueous metal cations at pH 2-7. Chemical
Geology, 151(1-4):215.
Vasquez, F.A., Heaviside, R., Tang, Z.J., and Taylor, J.S. 2006. Effect of free chlorine and chlora-
mines on lead release in a distribution system. Jour. AWWA, 98(2):144.
Xie, L., and Giammar, D.E. 2007. Equilibrium solubility and dissolution rate of the lead phosphate
chloropyromorphite. Environmental Science and Technology, 41(23):8050.
Yang, J., Mosby, D.E., Casteel, S.W., and Blanchar, R.W. 2001. Lead immobilization using phos-
phoric acid in a smelter-contaminated urban soil. Environmental Science and Technology,
35(17):3553.

ABBREVIATIONS
A specific surface area

AWWA American Water Works Association
BDL below detection limit

BET Brunauer Emmett and Teller adsorption isotherm
°C degrees Celsius
Ceff effluent concentration
Ceq predicted equilibrium concentration
Cinf influent concentration
cm2 square centimeters
Css steady-state concentration
CT total dissolved inorganic carbon concentration
C(t) concentration as a function of time
DIC dissolved inorganic carbon

ΔG Gibbs free energy of reaction
Ea activation energy
EDX energy dispersive X-ray analysis
f(ΔG) generic expression of function of Gibbs free energy effect on rate

FTIR Fourier transform infrared spectroscopy
ft/s feet per second
Gf°i Gibbs free energy of formation of species I

g/L grams per liter
IAP ion activity product

ICP-MS inductively coupled plasma mass spectrometry
k dissolution rate constant

k0 intrinsic dissolution rate constant
Ksp solubility product
L liters
LCR Lead and Copper Rule
M moles per liter

m2/g square meters per gram
mg C/L milligrams of carbon per liter
137

mg/L milligrams per liter

mg P/L milligrams of phosphorus per liter
mL milliliter
mM millimoles per liter
mol/min·m2 moles per minute per square meter
m/s meters per second
MWRA Massachusetts Water Resources Authority
μ chemical potential
μg/L micrograms per liter
μm micrometer
ng/L nanograms per liter

nH+ order of reaction with respect to H+
ni order of reaction with respect to species i
Ni stoichiometric coefficient of species i
nm nanometers
nM nanomoles per liter
NOM natural organic matter
ORP oxidation-reduction potential

ORS Octopole Reaction System
Ω empirical coefficient to account for non-elementary reactions
PAC project advisory committee

PCO2 partial pressure of carbon dioxide
PM project manager
ppb parts per billion
ppt parts per trillion
R ideal gas constant

rpm revolutions per minute
SEM scanning electron microscopy

SI saturation index
SIMS secondary ion mass spectrometry
s.t. stagnation time
T temperature
t time
tres hydraulic residence time
USEPA United States Environmental Protection Agency
XRD X-ray diffraction

6666 West Quincy Avenue, Denver, CO 80235-3098 USA
Water Chemistry Effects on Dissolution Rates of Lead Corrosion Products
P 303.347.6100 • F 303.734.0196 • www.WaterResearchFoundation.org
4064
1P-3C-4064-07/10-FP

Water Chemistry Effects On Dissolution Rates of Lead Corrosion Products

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Water Chemistry Effects On Dissolution Rates of Lead Corrosion Products

Caricato da

Copyright:

Formati disponibili

Water Chemistry Effects

Subject Area: Infrastructure

©2010 Water Research Foundation. ALL RIGHTS RESERVED

©2010 Water Research Foundation. ALL RIGHTS RESERVED

©2010 Water Research Foundation. ALL RIGHTS RESERVED

ALL RIGHTS RESERVED.

Printed in the U.S.A.

©2010 Water Research Foundation. ALL RIGHTS RESERVED

CHAPTER 2: REVIEW OF LEAD CARBONATE, PHOSPHATE, AND OXIDE

CHAPTER 3: DISSOLUTION RATES OF LEAD CORROSION PRODUCTS������������������������ 13

©2010 Water Research Foundation. ALL RIGHTS RESERVED

CHAPTER 4: RELEASE OF LEAD FROM PIPE SCALES������������������������������������������������������ 65

CHAPTER 5: SUMMARY AND CONCLUSIONS�������������������������������������������������������������������� 93

CHAPTER 6: RECOMMENDATIONS TO UTILITIES������������������������������������������������������������� 97

APPENDIX A: REVIEW OF LEAD CARBONATE, PHOSPHATE, AND OXIDE

APPENDIX B: EQUILIBRIUM CONSTANTS AND REACTIONS��������������������������������������� 113

APPENDIX C: RELATIONSHIP AMONG ALKALINITY, pH, AND DISSOLVED

APPENDIX D: TRACER STUDY OF CONTINUOUS-FLOW STIRRED REACTOR���������� 119

APPENDIX E: CALCULATING EQUILIBRIUM LEAD CONCENTRATIONS FOR

APPENDIX F: ORDER OF CONDUCTING EXPERIMENTS WITH LEAD PIPES ������������ 125

APPENDIX G: RESULTS OF EXPERIMENTS WITH LEAD SERVICE LINES������������������ 127

©2010 Water Research Foundation. ALL RIGHTS RESERVED

3.1 Solid phases studied in dissolution rate experiments......................................................... 14

3.2 Experimental variables in systematic investigation of dissolution rates........................... 19

3.3 Conditions studied for hydrocerussite dissolution............................................................. 22

3.4 Conditions and results of hydrocerussite dissolution experiments.................................... 23

3.5 Conditions studied for plattnerite dissolution.................................................................... 35

3.6 Conditions and results of plattnerite dissolution experiments........................................... 36

3.7 Conditions studied for hydroxylpyromorphite dissolution................................................ 49

3.8 Conditions and results of hydroxylpyromorphite dissolution experiments....................... 50

4.2 Factors evaluated in experiments with pipe reactors......................................................... 70

4.4 Dissolved lead divided by total lead concentration........................................................... 83

A.2 Summary of selected previous studies of carbonate dissolution rates............................. 109

A.3 Summary of selected previous studies of phosphate dissolution rates............................ 110

B.1 Equilibrium constants for aqueous species...................................................................... 114

B.2 Chemical potentials for various aqueous species............................................................. 115

B.3 Equilibrium constants for select reactions of aqueous species........................................ 115

B.4 Solubility product of select lead solids............................................................................ 116

C.1 Alkalinity for combinations of pH and DIC investigated................................................ 117

©2010 Water Research Foundation. ALL RIGHTS RESERVED

E.1 Equilibrium constants for aqueous species...................................................................... 122

G.1 Conditions and results of lead release experiments......................................................... 127

©2010 Water Research Foundation. ALL RIGHTS RESERVED

ES.1 Formation and stability of lead-containing corrosion products in the scales

1.1 Formation and stability of lead-containing corrosion products in the scales

1.2 Overview of research approach.......................................................................................... 3

2.1 Lead solubility as a function of pH..................................................................................... 8

3.1 XRD pattern of hydrocerussite and reference pattern....................................................... 14

3.2 Synthesized hydrocerussite at two different levels of magnification................................ 15

3.3 Synthesized cerussite........................................................................................................ 15

3.4 X-ray diffraction patterns of as-synthesized and acid-washed hydroxylpyromorphite.... 16

3.5 Acid-washed hydroxylpyromorhophite............................................................................ 16

3.6 Scanning electron micrograph of plattnerite..................................................................... 17

3.7 X-ray diffraction patterns of plattnerite............................................................................ 17

3.8 Flow-through reactor for measuring dissolution rates...................................................... 19

3.9 Illustration of the dissolved lead effluent from a flow-through reactor............................ 20

3.11 Dissolution rates of hydrocerussite as a function of pH with influent

3.12 Predicted equilibrium dissolved lead concentrations (lines) of 1 g/L hydrocerussite

©2010 Water Research Foundation. ALL RIGHTS RESERVED

3.15 Dissolution rates of hydrocerussite as a function of pH at a DIC = 0 mg/L with

3.16 Dissolution rates of hydrocerussite as a function of pH and dissolved inorganic

3.17 Powder X-ray diffraction of hydrocerussite reacted with an influent concentration

3.18 Scanning electron microscopy image of (A) hydrocerussite reacted at pH = 10

CHAPTER 3: DISSOLUTION RATES OF LEAD CORROSION PRODUCTS�� 13

CHAPTER 4: RELEASE OF LEAD FROM PIPE SCALES�� 65

CHAPTER 5: SUMMARY AND CONCLUSIONS�� 93

CHAPTER 6: RECOMMENDATIONS TO UTILITIES�� 97

APPENDIX B: EQUILIBRIUM CONSTANTS AND REACTIONS�� 113

APPENDIX D: TRACER STUDY OF CONTINUOUS-FLOW STIRRED REACTOR�� 119

APPENDIX F: ORDER OF CONDUCTING EXPERIMENTS WITH LEAD PIPES �� 125

APPENDIX G: RESULTS OF EXPERIMENTS WITH LEAD SERVICE LINES�� 127

3.39 Hydroxylpyromorphite after reacting with 2 mg Cl2/L of chloramine at pH 8.5............. 60